The New Stack Podcast

KubeCon 2023 is set to feature three hot topics, according to Taylor Dolezal from the Cloud Native Computing Foundation. Firstly, GenAI and Large Language Models (LLMs) are taking the spotlight, particularly regarding their security and integration with legacy infrastructure. Platform engineering is also on the rise, with over 25 sessions at KubeCon Chicago focusing on its definition and how it benefits internal product teams by fostering a culture of product proliferation. Lastly, WebAssembly is emerging as a significant topic, with a dedicated day during the conference week. It is maturing and finding its place, potentially complementing containers, especially in edge computing scenarios. Wasm allows for efficient data processing before data reaches the cloud, adding depth to architectural possibilities.Overall, these three trends are expected to dominate discussions and presentations at KubeCon NA 2023, offering insights into the future of cloud-native technology.See what came out of the last KubeCon event in Amsterdam earlier this year:AI Talk at KubeConDon’t Force Containers and Disrupt WorkflowsA Boring Kubernetes Release

Digital.ai, an AI-powered DevSecOps platform, serves large enterprises such as financial institutions, insurance companies, and gaming firms. The primary challenge faced by these clients is scaling their DevOps practices across vast organizations. They aim to combine modern development methodologies like agile DevOps with the need for speed and intimacy with end-users on a large scale.This episode features a discussion between Wing To of Digital.ai and TNS host Heather Joslyn about platform engineering and the role of AI in enhancing automation. It delves into the dilemma of whether increased code production and release frequency driven by DevOps practices are inherently beneficial. Additionally, it explores the emerging challenge of AI-assisted development and how large enterprises are striving to realize productivity gains across their organizations.Digital.ai is focused on incorporating AI into automation to assist developers in creating and delivering code while helping organizations derive more business value from their software in production. The company employs templates to capture and replicate key aspects of software delivery processes and uses AI to automate the rapid setup of developer environments and tooling. These efforts contribute to the concept of the internal developer platform, which consists of multiple toolsets for tasks like creating pipelines and setting up various components.Learn more from The New Stack about Platform Engineering, DevSecOps and Digital.ai:Platform Engineering Overview, News, and TrendsSRE vs. DevOps vs. Platform EngineeringMeet the New DevSecOps

Moving workloads to the cloud presents cost prediction challenges. Traditional setups with on-premises hardware offer predictability, but cloud costs are usage-based and granular. In this podcast episode, Matt Stellpflug, a senior FinOps specialist at ProsperOps, discusses the complexities of forecasting cloud expenses with TNS host Heather Joslyn.Cloud users face fluctuating costs due to continuous deployments and changing workloads. There are additional expenses for data access and transfer. Stellpflug emphasizes the importance of establishing reference workloads and benchmarks for accurate forecasting.Engineers play a vital role in FinOps initiatives since they ensure application availability and system integrity. Stellpflug suggests collaborating with engineering teams to identify essential metrics. He co-authored an "Engineer's Guide to Cloud Cost Optimization," highlighting the distinction between resource and rate optimization. Best practices involve addressing high-impact, low-risk areas first, engaging subject matter experts for complex issues, and maintaining momentum. This episode also provides further insights into implementing FinOps for effective cloud cost management.Learn more from The New Stack about FinOps and ProsperOps:FinOps Overview, News, and TrendsProsperOps Wants to Automate Your FinOps StrategyEngineer’s Guide to Cloud Cost Optimization: Manual DIY OptimizationEngineer’s Guide to Cloud Cost Optimization: Engineering Resources in the CloudEngineer’s Guide to Cloud Cost Optimization: Prioritize Cloud Rate Optimization

In her keynote address at the Linux Foundation's Open Source Summit Europe, Fatima Sarah Khalid emphasized that being an ally is more than just superficial gestures like wearing pronouns on badges or correctly pronouncing coworkers' names. True allyship involves taking meaningful actions to support and uplift individuals from underrepresented or marginalized backgrounds. This support is essential, not only in obvious ways but also in everyday interactions, which collectively create a more inclusive community.Open source communities typically lack diversity, with only a small percentage of women, non-binary contributors, and individuals from underrepresented backgrounds. Khalid stressed the importance of improving diversity and inclusion through various means, including using inclusive language, facilitating asynchronous communication to accommodate global contributors, and welcoming non-technical contributions such as documentation.Khalid also provided insights on making open source events more inclusive, like welcoming newcomers and marginalized groups, providing quiet spaces and enforcing a code of conduct, and partnering newcomers with mentors. Moreover, she highlighted GitLab's unique approach to allyship within the organization, including the Ally Lab, which pairs employees from different backgrounds to learn about and understand each other's experiences.To encourage the audience to embrace allyship, Khalid shared a set of commitments to keep in mind, such as educating oneself about the experiences of marginalized groups, speaking up against inappropriate behavior, using one's voice to amplify marginalized voices, donating to support such groups, and advocating for equity and justice through social networks and connections. She also shared real-life examples of allyship, illustrating how meaningful actions can create positive change in communities.Khalid's discussion with host Jennifer Riggins emphasizes the significance of meaningful, everyday actions to promote allyship in open source communities and organizations, ultimately contributing to a more diverse, inclusive, and equitable tech industry.Learn more from The New Stack about Open Source, Allyship, and GitLab:Embracing Open Source for Greater Business ImpactLeadership and Inclusion in the Open Source CommunityHow Implicit Bias Impacts Open Source Diversity and InclusionInvesting in the Next Generation of Tech Talent

In a recent conversation at the Open Source Summit in Bilbao, Spain, Gabriel Colombo, the General Manager of the Linux Foundation Europe and the Executive Director of the Fintech Open Source Foundation, discussed the potential impact of the Cyber Resilience Act (CRA) on the open source community. The conversation shed light on the challenges and opportunities that the CRA presents to open source and how individuals and organizations can respond.The conversation began by addressing the Cyber Resilience Act and its significance. Gabriel Colombo explained that while the Act is being touted as a measure to bolster cybersecurity and national security, it could have unintended consequences for the open source ecosystem, particularly in Europe. The Act, currently in the legislative process, aims to address cybersecurity concerns but could inadvertently hinder open source development and collaboration.Jim Zemlin, the Executive Director of the Linux Foundation, had previously mentioned the importance of forks in open source development, emphasizing that they are a healthy aspect of the ecosystem. However, Colombo pointed out that the CRA could create a sense of unease, as it might deter people and companies from participating in open source projects or using open source software due to potential legal liabilities.To grasp the implications of the CRA, Colombo explained some of the key provisions. The initial drafts of the Act proposed potential liability for individual developers, open source foundations, and package managers. This raised concerns about the open source supply chain's potential vulnerability and the distribution of liability.As the Act evolves, the liability landscape has shifted somewhat. Individual developers may not be held liable unless they consistently receive donations from commercial companies. However, for open source foundations, especially those accepting recurring donations from commercial entities, there remains a concern about potential liabilities and the need to conform to the CRA's requirements.Colombo emphasized that this issue isn't limited to Europe. It could impact the entire global open source ecosystem and affect the ability of European developers and small to medium-sized businesses to participate effectively.The conversation highlighted the challenges open source communities face when engaging with policymakers. Open source is not structured like traditional corporations or industry consortiums, making it more challenging to present a unified front. Additionally, the legislative process can be slow and complex, which may not align with the rapid pace of technology development.The lack of proactive engagement from the European Commission and the absence of open source communities in the initial consultations on the Act are concerning. The understanding of open source, its nuances, and the role it plays in the broader software supply chain appears limited within policy-making circles.What Can Be Done?Gabriel Colombo stressed the importance of awareness and education. It is vital for individuals, businesses, and open source foundations to understand the implications of the CRA. The Linux Foundation and other organizations have launched campaigns to provide information and resources to help stakeholders comprehend the Act's potential impact.Being vocal and advocating for open source within your network, organization, and through public affairs channels can also make a difference. Engagement with policymakers, especially as the Act progresses through the legislative process, is crucial. Colombo encouraged businesses to emphasize the significance of open source in their operations and supply chains, making policymakers aware of how the CRA might affect their activities.In the face of the Cyber Resilience Act, the open source community must unite and actively engage with policymakers. It's essential to educate and raise awareness about the potential impact of the Act and advocate for a balanced approach that strengthens cybersecurity without stifling open source innovation.The Act's development is ongoing, and there is time for stakeholders to make their voices heard. With a united effort, the open source community can help shape the legislation to ensure that open source remains vibrant and resilient in the face of evolving cybersecurity challenges.Learn more from The New Stack about open source and Linux Foundation Europe:At Open Source Summit: Introducing Linux Foundation EuropeMaking Europe's 'Romantic' Open Source World More PracticalEmbracing Open Source for Greater Business Impact

In this episode of The New Stack Makers podcast, Uma Daniel, a product manager at UST, discusses the current complexities in the global economy, marked by low unemployment except in the tech industry, high inflation, high interest rates, a volatile stock market, and the looming threat of recession. Amid these challenges, organizations are seeking ways to enhance their operational efficiency.Daniel introduces the concept of FinOps, which goes beyond just managing cloud costs. Instead, it focuses on leveraging the cloud to generate revenue. This represents a cultural shift in many organizations, emphasizing the need for a mindset change across different departments, including business, finance, and procurement.She dispels misconceptions, such as the belief that only certain teams should be involved in the FinOps process. Daniel stresses that it's a collaborative effort involving various teams, and it's best to adopt FinOps at the beginning of a cloud journey. Once an organization is already established in the cloud, implementing FinOps becomes more challenging.To foster collaboration, Daniel suggests identifying team members willing to champion FinOps and forming cross-functional teams to lead the initiative. Regular committee meetings and the establishment of generic policies, such as project budgets, help control cloud spending.This episode, hosted by Heather Joslyn, provides insights into how to initiate and implement a FinOps strategy and highlights common ways in which organizations waste cloud resources.Learn more from The New Stack about FinOps and UST:Cloud Cost-Unit Economics — A Modern Profitability ModelWhat Is FinOps? Understanding FinOps Best Practices for CloudVery Large Enterprises Need a Different Approach to FinOps

Since the release of OpenAI's ChatGPT-3 in late 2022, various industries have been actively exploring its applications. Madhukar Kumar, CMO of SingleStore, discussed his experiments with large language models (LLMs) in this podcast episode with TNS host Heather Joslyn. He mentioned a specific LLM called Gorilla, which is trained on APIs and can generate APIs based on specific tasks. Kumar also talked about SingleStore Now, an AI conference, where they plan to teach attendees how to build generative AI applications from scratch, focusing on enterprise applications.Kumar highlighted a limitation with current LLMs - they are "frozen in time" and cannot provide real-time information. To address this, a method called "retrieval augmented generation" (RAG) has emerged. SingleStore is using RAG to keep LLMs updated. In this approach, a user query is first matched with up-to-date enterprise data to provide context, and then the LLM is tasked with generating answers based on this context. This method aims to prevent the generation of factually incorrect responses and relies on storing data as vectors for efficient real-time processing, which SingleStore enables.This strategy ensures that LLMs can provide current and contextually accurate information, making AI applications more reliable and responsive for enterprises.Learn more from The New Stack about LLMs and SingleStore:Top 5 Large Language Models and How to Use Them EffectivelyUsing ChatGPT for Questions Specific to Your Company Data6 Reasons Private LLMs Are Key for Enterprises

Observability in multi-cloud environments is becoming increasingly complex, as highlighted by Martin Mao, CEO and co-founder of Chronosphere. This challenge has two main components: a rise in customer-facing incidents, which demand significant engineering time for debugging, and the ineffectiveness and high cost of existing tools. These issues are creating a problematic return on investment for the industry.Mao discussed these observability challenges on The New Stack Makers podcast with host Heather Joslyn, emphasizing the need to help teams prioritize alerts and encouraging a shift left approach for security responsibility among developers. With the adoption of distributed cloud architectures, organizations are not only dealing with a surge in data but also facing a cultural shift towards DevOps, where developers are expected to be more accountable for their software in production.Historically, operations teams handled software in production, but in the cloud-native world, developers must take on these responsibilities themselves. Many current observability tools were designed for centralized operations teams, which creates a gap in addressing developer needs.Mao suggests that cloud-native observability tools should empower developers to run and maintain their software in production, providing insights into the complex environments they work in. Moreover, observability tools can assist developers in understanding the intricacies of their software, such as its dependencies and operational aspects.To streamline the data obtained from observability efforts and manage costs, Chronosphere introduced the "Observability Data Optimization Cycle." This framework starts with establishing centralized governance to set budgets for teams generating data. The goal is to optimize data usage to extract value without incurring unnecessary costs. This approach applies financial operations (FinOps) concepts to the observability space, helping organizations tackle the challenges of cloud-native observability.Learn more from The New Stack about Observability and Chronosphere:Observability Overview, News and Trends4 Key Observability Best PracticesTop Ways to Reduce Your Observability CostsTop 4 Factors for Cloud Native Observability Tool Selection

Platform engineering is gaining prominence due to the need for faster application deployment, which directly impacts business velocity. Valentina Alaria, Senior Director of Product at VMware, emphasizes that not all organizations pursuing platform engineering have the same goals, context, or pain points. They tailor solutions to each organization's specific needs. Some focus on rapid onboarding for junior developers, while others aim to reduce complexity, friction, and support larger development teams with fewer operational staff.Platform engineering aims to streamline collaboration between developers and operations engineers. Developers want portable code and the ability to focus on coding without worrying about production requirements. Operations engineers and platform teams seek a seamless environment for deploying applications in different contexts.Successful platform engineering initiatives involve strong collaboration models, fostering a cooperative approach rather than a siloed one. The goal is to create applications and value for the organization by facilitating effective interaction between developers and operations engineers.This podcast episode, hosted by Alex Williams of TNS, also delves into VMware Tanzu's latest tools for supporting platform engineering.Learn more from The New Stack about platform engineering and VMware Tanzu:Platform Engineering Overview, News and Trends6 Patterns for Platform Engineering SuccessA Guide to Open Source Platform EngineeringStreamline Platform Engineering with Kubernetes

ByConity is an open source project that emerged from ByteDance's use of Clickhouse, an open-source database system, to address their growing data volume. ByConity focuses on enhancing the separation of compute and storage, improving multitenancy support, and optimizing query performance in cloud-native environments.ByteDance's Vini Jaiswal, a principle developer advocate at the parent company of TikTok, highlights the power of open source in fostering innovation and collaboration. She shares her personal experience of leveraging open source to solve problems quickly and efficiently. She emphasizes the importance of getting involved in open source, even for those who might be hesitant, and suggests starting by identifying a pain point and making small contributions.ByConity's architecture, which separates compute and storage, offers benefits like preventing data lake corruption, read and write separation, elasticity, and scalability. Jaiswal also mentions her previous experience with open source during her time at CitiBank, where she realized how open source accelerated digital transformations.Throughout the conversation, Jaiswal underscores the strength of open source communities in collectively addressing challenges. She encourages listeners to embrace open source and start contributing, emphasizing how even small contributions can lead to significant impacts over time.The episode also delves into Jaiswal's involvement with other open source projects, such as PyTorch, and explores the intersection of open source and generative AI.Learn more from The New Stack about open source and cloud native environments:What Is 'Cloud Native' (and Why Does It Matter)?Cloud Native Ecosystem News and ResourcesHow to Build an Open Source Community

Along with discussing the emergence and ascension of platform engineering in this episode, we also discuss the role that Humanitec plays in helping organizations establish platforms for developers, as well as Backstage, a popular open source internal developer platform that was developed by Spotify for its own developers.An IDP, our guest Kaspar Von Grünberg explained, is a standardized interface for developers to build applications using a golden path of vetted tools and libraries, allowing for a high degree of efficiency for both the developers themselves as well as the engineers who are supporting the developers. They can include an integration and delivery plane, a continuous integration registry, a platform orchestrator, observability tools and a resource plane."How you're consuming this is a little bit up to the individual preference of the user, and what the platform team has configured for you. So we're seeing some teams like to use a user interface and some teams like to use code based interactions," Von Grünberg explained.In some ways, a IDP is reminiscent of the platform-as-a-service packages of a decade ago. They also were designed to help developer efficiency, though devs chafed at the limited number of tools they were allowed to use in these walled gardens. That was a mistake, Von Grünberg said.Those platforms required developers to use a small set of pre-defined times."We don't want to get back to those times, which is why we want to provide sensible defaults," Von Grünberg said. A good IDP will provide developers with "golden paths" or "paved roads" as Netflix calls them."Developers can stay on those paths if they want," Von Grünberg said. They can enjoy the security default and service-level agreements (SLAs) from the engineers. But developers are also free to leave the path and make low-level configurations on their own as well."Good platform engineering is never about covering all the use cases," he said.Learn more from The New Stack about platform engineering and Humanitec:Platform Engineering Overview, News, and TrendsHow to Pave Golden Paths That Actually Go SomewhereBuild Your IDP at Light Speed with a Platform Reference Architecture

In this episode of The New Stack Makers, technologist and author John Willis emphasized caution when considering AI solutions from vendors. He advised against blindly following vendor recommendations for "one-size-fits-all" AI products, likening it to discouraging learning Java in the past in favor of purchasing a product.Willis stressed that DevOps serves as an example of how human expertise, not just products, solves problems. He urged C-level executives to first understand AI's intricacies and then make informed purchasing decisions, suggesting a "DevOps redo" to encourage experimentation and collaboration, similar to the early days of the DevOps movement.Willis highlighted that early adopters of DevOps, like successful banks, heavily invested in developing their human capital. He cautioned against hasty product purchases, as the AI landscape is rife with startups that may quickly disappear or be acquired by larger companies.Instead, Willis advocated for educating teams on effective data management techniques, including retrieval augmentation, to fine-tune large language models. He emphasized the need for data cleansing to build robust data pipelines and prevent LLMs from generating undesirable code or sensitive information.According to Willis, the process becomes enjoyable when done correctly, especially for companies using LLMs at scale with retrieval augmentation. To ensure success, he suggested adding governance and structure, including content moderation and red-teaming of data, which vendors may not prioritize in their offerings.Learn more from The New Stack about DevOps and AI:AIOps: Is DevOps Ready for an Infusion of Artificial Intelligence?How to Build a DevOps Engineer in Just 6 MonthsPower up Your DevOps Workflow with AI and ChatGPT

Deliveroo, a prominent food delivery company, relies on Apache Flink, a distributed processing engine, to enhance its three-sided marketplace, connecting delivery drivers, restaurants, and customers. Seeking to improve real-time data streaming and gain insights into customer behavior, Deliveroo transitioned to Flink, comparing it to alternatives like Apache Spark and Kafka Streams. Flink, with feature parity to their previous platform, offered stability and scalability. They initially experimented with Flink on Kubernetes but turned to the Amazon Managed Service for Flink (MSF) for enhanced support and maintenance.Engineers from Deliveroo, Felix Angell and Duc Anh Khu, emphasized the need for flexibility in data modeling to accommodate their fast-paced product development. However, flexibility can be complex, often requiring data model adjustments. They expressed the desire for a self-serve configuration feature in MSF, allowing easy customization of low-level settings and auto-scaling based on application metrics. This move to Flink and MSF has empowered Deliveroo to focus on core responsibilities like continuous integration and delivery while efficiently managing their data processing needs.Learn more from The New Stack about Apache Flink and AWS:Kinesis, Kafka and Amazon Managed Service for Apache FlinkApache Flink for Real Time Data AnalysisApache Flink for Unbounded Data Streams

Over the past five to ten years, the testing of microservices has seen significant growth. This surge in testing can be attributed to the increasing adoption of microservices and Kubernetes, which signify a shift away from monolithic application architectures. Bruno Lopes, a leader at Kubernetes company incubator Kubeshop, noted this trend. Kubeshop has initiated six Kubernetes projects, including TestKube, a Kubernetes native testing framework led by Lopes.This rise in testing is making it more accessible to a wider audience and is enhancing the developer experience through automation. Developers now have more time to focus on innovation rather than manual testing. However, there is often a disconnect between development and testing, as developers move quickly, outpacing organizational adaptation to modern testing methods.Lopes emphasized the importance of testing before production deployment and advocated for creating production-resembling testing environments that allow for rapid deployment without waiting for manual tests. This approach is particularly critical for Site Reliability Engineering (SRE) teams who need to respond quickly to issues and minimize downtime for customers. In some cases, it's necessary to run tests within Kubernetes itself, a concept that may take time for companies to fully embrace as the developer experience continues to improve.Learn more from The New Stack about Kubernetes, Testing and TestKube:Testkube: A Cloud Native Testing Framework for KubernetesTop 5 Challenges in Modern Kubernetes TestingWhy You Should Start Testing in the Cloud Native Way

Apache Flink is an open-source framework and distributed processing engine designed for data analytics. It excels at handling tasks such as data joins, aggregations, and ETL (Extract, Transform, Load) operations. Moreover, it supports advanced real-time techniques like complex event processing.In this episode, Deepthi Mohan and Nagesh Honnalii from AWS discussed Apache Flink and the Amazon Managed Service for Apache Flink (MSF) with our host, Alex Williams. MSF is a service that caters to customers with varying infrastructure preferences. Some prefer complete control, while others want AWS to handle all infrastructure-related aspects.Use cases for MSF can be grouped into three categories. First, there's streaming ETL, which involves tasks like log aggregation for later auditing. Second, it supports real-time analytics, enabling customers to create dashboards for tasks like fraud detection. Third, it handles complex event processing, where data from multiple sources is joined and aggregated to extract meaningful insights.The origins of MSF trace back to the evolution of real-time data services within AWS. In 2013, AWS introduced Amazon Kinesis, while the open-source community developed Apache Kafka. These services paved the way for MSF by highlighting the need for real-time data processing.To provide more flexibility, AWS launched Kinesis Data Analytics in 2016, allowing customers to write code in JVM-based languages like Java and Scala. In 2018, AWS decided to incorporate Apache Flink into its Kinesis Data Analytics offering, leading to the birth of MSF.Today, thousands of customers use MSF, and AWS continues to enhance its offerings in the real-time data processing space, including the launch of Amazon MSK (Managed Streaming for Apache Kafka). To align with its foundation on Flink, AWS rebranded Kinesis Data Analytics for Apache Flink to Amazon Managed Service for Apache Flink, making it clearer for customers.Learn more from The New Stack about AWS and Apache Flink:Apache Flink for Real Time Data AnalysisApache Flink for Unbounded Data Streams3 Reasons Why You Need Apache Flink for Stream Processing

Modern developer conferences like the upcoming Infobip Shift Conference in Croatia are centered around themes. At this particular event for developers, you can expect a lot of focus to be on the developer experience and artificial intelligence (AI).Ivan Burazin, Chief Development Experience Officer at InfoBip, joined us on the show and emphasizes that developers spend a substantial portion of their time not coding, often losing 50 to 70% of their productive hours to non-coding activities, such as setting up environments, running tests, and building code. This highlights the importance of improving the developer experience to enhance productivity.The developer experience has both internal and external dimensions. Externally, it impacts customer experience, while internally, it influences development velocity. A better developer experience translates to faster and more efficient coding.The Shift Conference will feature talks on six stages, one of which will focus on the developer experience, addressing its internal and external aspects. Additionally, AI will take center stage at another segment of the conference.Although there may not be an abundance of true AI experts taking the stage, the focus will be on how individuals and companies can leverage AI to create products and services. It's recognized that AI will play a pivotal role in the future of every industry, and the conference aims to explore practical applications and strategies for integrating AI into various businesses.Overall, the Shift Conference aims to address the challenges developers face in optimizing their productivity and explore the growing importance of AI in shaping the future of businesses and products.Learn more from The New Stack about the developer experience and InfoBip Shift:7 Principles and 10 Tactics to Make You a 10x DeveloperThe Challenges of Marketing Software Tools to DevelopersA Guide to Better Developer Experience

This episode delves into Apache Flink, a versatile platform for executing both batch and real-time streaming data analysis tasks. This session marks the beginning of a three-part series unveiling Amazon Web Services' (AWS) new managed service built on Flink. Future episodes will explore this service in detail and examine customer experiences.The podcast features insights from Danny Cranmer, a principal engineer at AWS and an Apache Flink PMC and Committer, along with Hong Teoh, a software development engineer at AWS.Flink stands out as a high-level framework for defining data analytics jobs, accommodating both batch and streaming data sets. It offers APIs for building analysis jobs in various languages, including Java, Python, and SQL. Flink also provides a distributed job execution engine with fault tolerance and horizontal scaling capabilities.One prominent use case is Extract-Transform-Load (ETL), where raw data is swiftly processed for specific workloads. Flink excels in delivering low-latency transformations for unbounded data streams. Additionally, Flink supports event-driven applications, responding immediately to triggers such as user requests for weather data.Flink ensures exactly-once processing, critical for scenarios like financial transactions. It employs checkpoints to maintain data integrity in case of node failures.The podcast also touches on AWS's role in supporting the open-source Flink project and the future outlook for this powerful data processing framework.Learn more from The New Stack about Apache Flink:3 Reasons Why You Need Apache Flink for Stream ProcessingApache Flink for Unbounded Data Streams8 Real-Time Data Best Practices

In an interview with The New Stack, renowned technologist Adrian Cockcroft discussed the process of fine-tuning Large Language Models (LLMs) through prompt engineering. Cockcroft, known for his roles at Netflix and Amazon Web Services, explained how to obtain tailored programming advice from an LLM. By crafting specific prompts like asking the model to provide code in the style of a certain expert programmer, such as Java's James Gosling, users can guide the AI's output.Prompt engineering involves setting up conversations to bias the AI's responses. These prompts are becoming more advanced with plugins and loaded information that shape the model's behavior before use. Cockcroft highlighted the concept of fine-tuning, where models are adapted beyond what a prompt can contain. Companies are incorporating vast amounts of their internal data, like wiki pages and corporate documents, to train the model to understand their specific domain and processes.Cockcroft pointed out the efficacy of ChatGPT within certain tasks, illustrated by his experience using it for data analysis and programming assistance. He also discussed the growing need for improved results from LLMs, which has led to the demand for vector databases. These databases store word meanings as vectors with associated weights, enabling fuzzy matching for enhanced information retrieval from LLMs. In essence, Cockcroft emphasized the multifaceted process of shaping and optimizing LLMs through prompt engineering and fine-tuning, reflecting the evolving landscape of AI-human interactions.Learn more from The New Stack about LLMs and Prompt Engineering:Top 5 Large Language Models and How to Use Them EffectivelyThe Pros (And Con) of Customizing Large Language ModelsPrompt Engineering: Get LLMs to Generate the Content You WantDeveloper Tips in AI Prompt Engineering

TechWorld with Nana is one of the most popular resources for people looking to get into or progress a DevOps career. Nana Janashia, the creator of TechWorld with Nana, is a DevOps trainer and consultant who joined us to discuss why DevOps is needed now more than ever and how this is the perfect time to begin a career in DevOps.Host Alex Williams and Nana go over the key concepts of DevOps. Then they talk about how the complexity of tools can sidetrack and complicate the learning process for those new to DevOps and why focusing on concepts rather than tools the way to go. Before wrapping up the conversation, they even talk about the best ways for people to get involved who are new to DevOps.Nana's journey into DevOps commenced during her time as an engineer in Austria, where she began exploring Kubernetes. As inquiries from colleagues poured in, she recognized her knack for demystifying complex topics, catalyzing her passion for teaching. Viewers attest to switching to DevOps careers after watching her videos.Throughout the conversation, we learned how people can discover the world of DevOps through TechWorld with Nana as an expert guide. With a large YouTube audience, online courses, workshops, and corporate training, Nana has empowered countless individuals in advancing their DevOps expertise. The six-month boot camps from TechWorld with Nana encompass a comprehensive curriculum, starting with fundamentals and culminating in hands-on programming abilities, Python automation, configuration management, and Prometheus-based monitoring.Nana underscores that DevOps, still a relatively nascent profession, suffers from role ambiguity both among engineers and within companies aspiring to implement it. This confusion stems from differing workflows and environments when engineers switch jobs. Nana's insights bring clarity to these challenges, acknowledging the evolving chaos of the DevOps culture and its driving force for innovation in managing intricate distributed technologies.Learn more about DevOps from TNS, Roadmap (our sister site), and TechWorld with Nana:TechWorld with Nana - DevOps BootcampTechWorld with Nana - DevSecOps BootcampDevOps Learning RoadmapDevOps News, Trends, and Analysis

Explore the complex intersection of AI and open source with insights from experts in this illuminating discussion. Amanda Brock, CEO of OpenUK, reveals the challenges in labeling AI as open source amidst legal ambiguities. The dialogue, led by TNS host Alex Williams, delves into the evolution of open source licensing, its departure from traditional models, and the complications arising from applying open source principles to AI, which encompasses sensitive data governed by privacy laws.The focus turns to "Llama 2," a contentious example where Meta labeled their language model as open source, sparking confusion. Notable guests Erica Brescia, Managing Director at Redpoint Ventures, and Steven Vaughan-Nichols, founder of Open Source Watch, weigh in on this topic. Brock emphasizes that AI's complexity prevents it from aligning with the Open Source Definition, necessitating a clear distinction between open innovation and open source.Amidst these debates, the Open Source Initiative (OSI) is crafting a new definition tailored for AI, sparking anticipation and discussion about its implications. The necessity for an evolved understanding of open source and its licenses is underscored, as the rapid evolution of technology challenges established norms. The journey concludes with reflections on vendors transitioning from open source licenses to Server Side Public License (SSPL) due to cloud-related considerations, raising questions about the future of open source in a dynamically changing tech landscape.Learn more from The New Stack about open source and AI:Open Source May Yet Eat Google's and OpenAI's AI LunchOpen Source Movement Emerging in AI To Counter GreedHow AI Can Learn from the Struggles of Open Source

Discover how large language models and generative AI are revolutionizing DevOps with PromptOps. The company, initially known as CtrlStack, introduces its unique process engine that comprehends human requests, reads knowledge bases, and generates code on the fly to accomplish tasks. Dev Nag, the CEO, explains how PromptOps saves users time and money by automating routine operations in this podcast episode with The New Stack.Dev Nag is joined by GK Brar, PromptOps' founding engineer, and our host Joab Jackson as they delve into the concept of generative AI and its potential benefits for DevOps. Traditionally, DevOps tasks often involve repetitive troubleshooting and reporting, making automation essential. PromptOps specializes in intent matching, understanding nuanced requests and providing the right solutions.Notably, PromptOps employs generative AI offline to prepare for automating common actions and enhancing the user experience. Unlike others, PromptOps aims beyond simple enhancements. It aspires to transform the entire DevOps landscape by leveraging this groundbreaking technology.Tune in to the podcast to gain deeper insights into this transformative approach that PromptOps brings to DevOps thanks to the power and possibilities of generative AI.Learn more from The New Stack about DevOps and PromptOps:DevOps News, Trends, Analysis and ResourcesHow to Use ChatGPT for IT Security AuditWhat We Learned from Building a Chatbot

In this episode, Matt Butcher, CEO of Fermyon Technologies, discusses the potential impact of the component model on WebAssembly (Wasm) and its integration into the cloud-native landscape. WebAssembly is a binary instruction format enabling code to run anywhere, written in developers' preferred languages. The component model aims to provide a common way for WebAssembly libraries to express their needs and connect with other modules, reducing the barriers and maintenance of existing libraries. Butcher believes this model could be a game changer, allowing new languages to compile WebAssembly and utilize existing libraries seamlessly.WebAssembly also shows promise in delivering on the long-awaited potential of serverless computing. Unlike traditional virtual machines and containers, WebAssembly boasts a rapid startup time and addresses various developer challenges. Butcher states that developers have been eagerly waiting for a platform with these characteristics, hinting at a potential resurgence of serverless. He clarifies that WebAssembly is not a "Kubernetes killer" but can coexist with container technologies, evident from the Kubernetes ecosystem's interest in supporting WebAssembly.The episode explores further developments in WebAssembly and its potential to play a central role in the cloud-native ecosystem.Learn more from The New Stack about WebAssembly and Fermyon Technologies:WebAssembly Overview, News, and TrendsWebAssembly vs. KubernetesFermyon Cloud: Save Your WebAssembly Serverless Data Locally

Building and deploying applications in the cloud offers significant advantages, primarily driven by the scalability it provides. Developers appreciate the speed and ease with which cloud-based infrastructure can be set up, allowing them to scale rapidly as long as they have the necessary resources. However, the very scale that makes cloud computing attractive also poses serious risks.The risk lies in the potential for developers to make mistakes in application building, which can lead to widespread consequences when deployed at scale. Cloud-focused attacks have seen a significant increase, tripling from 2021 to 2022, as reported in the Cloud Risk Report by Crowdstrike.The challenges in securing the cloud are exacerbated by its relative novelty, with organizations still learning about its intricacies. The newer generation of adversaries is adept at exploiting cloud weaknesses and finding ways to attack multiple systems simultaneously. Cultural issues within organizations, such as the tension between security professionals and developers, can further complicate cloud protection.To safeguard cloud infrastructure, best practices include adopting the principle of least privilege, regularly evaluating access rights, and avoiding hard-coding credentials. Ongoing hygiene and assessments are crucial in ensuring that access levels are appropriate and minimizing risks of cloud-focused attacks.Overall, understanding and addressing the risks associated with cloud deployments are vital as cloud-native adversaries grow increasingly sophisticated. Implementing proper security measures, along with staying up-to-date on runtime security and avoiding misconfigurations, are essential in safeguarding cloud-based applications and data.Elia Zaitsev of CrowdStrike joined TNS host Heather Joslyn for this conversation on the heels of the release of their Cloud Risk Report.Learn more from The New Stack about cloud security and CrowdStrike:Cloud-Focused Attacks Growing More Frequent, More Brazen5 Best Practices for DevSecOps Teams to Ensure ComplianceWhat Is DevSecOps?

In this episode of The New Stack Makers, Purnima Padmanabhan, a senior vice president at VMware, discusses three common mistakes organizations make when trying to move faster in meeting customer needs. The first mistake is equating application modernization with solely moving to the cloud, often resulting in a mere lift and shift of applications, without reaping the full benefits. The second mistake is a lack of automation, particularly in operations, which hinders the development process's speed. The third mistake involves adding unnecessary complexity by adopting new technologies or procedures, which slows down developers.As a solution, Padmanabhan introduces the concept of platform engineering, which not only accelerates development but also reduces toil for operations engineers and architects. However, many organizations struggle with implementing it effectively, as they often approach platform engineering in fragmented ways, investing in separate components without fully connecting them.To succeed in adopting platform engineering, Padmanabhan emphasizes the need for a mindset shift. The platform team must treat platform engineering as a continuously evolving product rather than a one-time delivery, ensuring that service-level agreements are continuously met, and regularly updating and improving features and velocity. The episode discusses the benefits of a well-implemented "golden path" for entire organizations and provides insights on how to start a platform engineering team.Learn more from The New Stack about Platform Engineering and VMware:Platform Engineering Overview, News and TrendsPlatform Engineers: Developers Are Your CustomersOpen Source Platform Engineering: A Decade of Cloud Foundry

In this episode of The New Stack Makers, Peter Klimek, director of technology in the Office of the CTO at Imperva, discusses the vulnerability of business logic in a distributed, cloud-native environment. Business logic refers to the rules and processes that govern how applications function and how users interact with them and other systems. Klimek highlights the increasing attacks on APIs that exploit business logic vulnerabilities, with 17% of attacks on APIs in 2022 coming from malicious bots abusing business logic.The attacks on business logic take various forms, including credential stuffing attacks, carding (testing stolen credit cards), and newer forms like influence fraud, where algorithms are manipulated to deceive platforms and users. Klimek emphasizes that protecting business logic requires a cross-functional approach involving developers, operations engineers, security, and fraud teams.To enhance business logic security, Klimek recommends conducting a threat modeling exercise within the organization, which helps identify potential risk vectors. Additionally, he suggests referring to the Open Web Application Security Project (OWASP) website's list of automated threats as a checklist during the exercise.Ultimately, safeguarding business logic is crucial in securing cloud-native environments, and collaboration among various teams is essential to effectively mitigate potential threats and attacks.More from The New Stack, Imperva, and Peter Klimek:Why Your APIs Aren’t Safe — and What to Do about ItZero-Day Vulnerabilities Can Teach Us About Supply-Chain SecurityGraphQL APIs: Greater Flexibility Breeds New Security Woes

In this episode of The New Stack Makers podcast, the focus is on the challenges of handling unstructured data in today's data-rich world and the potential solutions offered by vector databases and vector searches. The use of relational databases is limited when dealing with text, images, and voice data, which makes it difficult to uncover meaningful relationships between different data points.Vector databases, which facilitate vector searches, have become increasingly popular for addressing this issue. They allow organizations to store, search, and index data that would be challenging to manage in traditional databases. Semantic search and Large Language Models have sparked interest in vector databases, providing developers with new possibilities.Beyond standard applications like information search and recommendation bots, vector searches have also proven useful in combating copyright infringement. Social media companies like Facebook have pioneered this approach by using vectors to check copyrighted media uploads.Vector databases excel at finding similarities between data objects, as they operate in vector spaces and perform approximate nearest neighbor searches, sacrificing a bit of accuracy for increased efficiency. However, developers need to understand their specific use cases and the scale of their applications to make the most of vector databases and search.Frank Liu, the director of operations at Zilliz, advised listeners to educate themselves about vector databases, vector search, and machine learning to leverage the existing ecosystem of tools effectively. One notable indexing strategy for vectors is Hierarchical Navigable Small Worlds (HNSW), a graph-based algorithm created by Yury Malkov, a distinguished software engineer at VerSE Innovation who also joined us along with Nils Reimers of Cohere.It's crucial to view vector databases and search as additional tools in the developer's toolbox rather than replacements for existing database management systems or document databases. The ultimate goal is to build applications focused on user satisfaction, not just optimizing clicks. To delve deeper into the topic and explore the gaps in current tooling, check out the full episode.Listen on PoduramaLearn more about vector databases at thenewstack.ioVector Databases: What Devs Need to Know about How They WorkVector Primer: Understand the Lingua Franca of Generative AIHow Large Language Models Fuel the Rise of Vector Databases

Sargun Kaur, co-founder of Byteboard, aims to revolutionize the tech interview process, which she believes is flawed and ineffective. In an interview with The New Stack for our Tech Founder Odyssey podcast series, Kaur compared assessing technical skills during interviews to evaluating the abilities of basketball star Steph Curry by asking him to draw plays on a whiteboard instead of watching him perform on the court. Kaur, a former employee of Symantec and Google, became motivated to change the interview process after a talented engineer she had coached failed a Google interview due to its impractical format.Kaur believes that traditional tech interviews overly emphasize theoretical questions that do not reflect real-world software engineering tasks. This not only limits the talent pool but also leads to mis-hires, where approximately one in four new employees is unsuitable for their roles or teams. To address these issues, Kaur co-founded Byteboard in 2018 with Nicole Hardson-Hurley, another former Google employee. Byteboard offers project-based technical interviews, adopted by companies like Dropbox, Lyft, and Robinhood, to enhance the efficiency and fairness of their hiring processes. In recognition of their work, Kaur and Hardson-Hurley received Forbes magazine's "30 Under 30" award for enterprise technology.Kaur's journey into the tech industry was unexpected, considering her initial disinterest in her father's software engineering career. However, exposure to programming and shadowing a female engineer at Microsoft sparked her curiosity, leading her to study computer science at the University of California, Berkeley. Overcoming initial challenges as a minority in the field, Kaur eventually joined Google as an engineer, content with the work environment and mentorship she received. However, her dissatisfaction with the interview process prompted her to apply to Google's Area 120 project incubator, leading to the creation of Byteboard. Kaur's experience with Byteboard's development and growth taught her valuable lessons about entrepreneurship, the power of founders in fundraising meetings, and the potential impact of AI on tech hiring processes.Check out more episodes in The Tech Founder Odyssey series:A Lifelong ‘Maker’ Tackles a Developer Onboarding ProblemHow Teleport’s Leader Transitioned from Engineer to CEOHow 2 Founders Sold Their Startup to Aqua Security in a Year

Shanea Leven, co-founder and CEO of CodeSee, shared her journey as a tech founder in an episode of the Tech Founder Odyssey podcast series. Despite coming to programming later than many of her peers, Leven always had a creative spark and a passion for making things. She initially pursued fashion design but taught herself programming in college and co-founded a company building custom websites for book authors. This experience eventually led her to a job at Google, where she worked in product development.While at Google, Leven realized the challenge of deciphering legacy code and onboarding developers to it. Inspired by a presentation by Bret Victor, she came up with the idea for CodeSee—a developer platform that helps teams understand and review code bases more effectively. She started working on CodeSee in 2019 as a side project, but it soon received venture capital funding, allowing her to quit her job and focus on the startup full-time.Leven candidly discussed the challenges of juggling a day job and a startup, particularly after receiving funding. She also shared advice on raising money from venture capitalists and building a company culture.Listen to the full episode and check out more installments from The Tech Founder Odyssey.How Teleport’s Leader Transitioned from Engineer to CEOHow 2 Founders Sold Their Startup to Aqua Security in a YearHow Solvo’s Co-Founder Got the ‘Guts’ to Be an Entrepreneur

In deploying cloud-native sustainable foundation AI models, there are five key steps outlined by Huamin Chen, an R&D professional at Red Hat's Office of the CTO. The first two steps involve using containers and Kubernetes to manage workloads and deploy them across a distributed infrastructure. Chen suggests employing PyTorch for programming and Jupyter Notebooks for debugging and evaluation, with Docker community files proving effective for containerizing workloads.The third step focuses on measurement and highlights the use of Prometheus, an open-source tool for event monitoring and alerting. Prometheus enables developers to gather metrics and analyze the correlation between foundation models and runtime environments.Analytics, the fourth step, involves leveraging existing analytics while establishing guidelines and benchmarks to assess energy usage and performance metrics. Chen emphasizes the need to challenge assumptions regarding energy consumption and model performance.Finally, the fifth step entails taking action based on the insights gained from analytics. By optimizing energy profiles for foundation models, the goal is to achieve greater energy efficiency, benefitting the community, society, and the environment.Chen underscores the significance of this optimization for a more sustainable future.Learn more at thenewstack.ioPyTorch Takes AI/ML Back to Its Research, Open Source RootsPyTorch Lightning and the Future of Open Source AIJupyter Notebooks: The Web-Based Dev Tool You've Been SeekingKnow the Hidden Costs of DIY Prometheus

The concept of a software bill of materials (SBOM) aims to provide consumers with information about the components inside a software, enabling better assessment of potential security issues. Justin Hutchings, Senior Director of Product Management at GitHub, emphasizes the importance of SBOMs and their potential to facilitate patching without relying solely on the vendor. He spoke with Alex Williams in this episode of The New Stack Makers.Creating a comprehensive SBOM poses challenges. Each software package is unique, such as an Android application that combines the developer's code with numerous open-source dependencies obtained through Maven packages. The SBOM should ideally serve as a machine-readable inventory of all these dependencies, enabling developers to evaluate their security.Hutchings notes that many SBOMs fall short in being fully machine-readable, and the vulnerability landscape is even more problematic. To achieve the standards Hutchings envisions, several actions are necessary. For instance, certain programming languages make it difficult to inspect build contents, while the lack of a centralized distribution point for dependencies in languages like C and C++ complicates the enumeration and standardization of machine-readable names and versions. Addressing these issues across the entire software supply chain is imperative.SBOMs hold potential for enhancing software security, but the current state of implementation and machine-readability needs improvement, particularly concerning diverse programming languages and dependency management.Learn more at thenewstack.ioCreating a 'Minimum Elements' SBOM Document in 5 MinutesEnhance Your SBOM Success with SLSAHow to Create a Software Bill of Materials

Angel Diaz, Vice President of Technology, Capabilities, and Innovation at Discover Financial Services, spoke with TNS Host Alex Williams at the Open Source Summit in Vancouver, BC. Diaz emphasizes the importance of learning and collaboration among software engineers. He leads The Discover Technology Academy, a community of 15,000 engineers, which he describes as a place where craftsmen come together rather than an ivory tower institution.Developers and engineers at Discover define and develop processes for software development. They start their journey by contributing atomic elements of knowledge, such as articles, blogs, videos, and tutorials, and then democratize that knowledge. Open source principles, communities, guilds, and established practices play a vital role in their work and discovery process.Discover's developer experience revolves around the concept of the golden path, which goes beyond consuming content and includes aspects like code, automation, and setting up development environments. Pair programming and a cultural approach to learning are also incorporated into Discover's talent system.Diaz highlights that Discover's work extends beyond their financial services company, as they share their knowledge and open source work with the external community through platforms like technology.discovered.com. This enables engineers to gain merit badges, such as maintainers or contributors, and showcase their expertise on professional platforms like LinkedIn.Learn more at thenewstack.ioThe Future of Developer CareersPlatform Engineer vs Software EngineerHow Donating Open Source Code Can Advance Your Career

The Linux Foundation's Open Source Security Foundation (OSSF) is addressing the challenge of timely software component updates to prevent security vulnerabilities like Log4J. In an interview with Alex Williams of The New Stack at the Open Source Summit in Vancouver, Omkhar Arasaratnam, the new general manager of OSSF, and Brian Behlendorf, CTO of OSSF, discuss the importance of making software secure from the start and the need for rapid response when vulnerabilities occur. In this conversation, they highlight the significance of Software Bill of Materials (SBOMs), which provide a complete list of software components and supply chain relationships. SBOMs offer data that can aid decision-making and enable reputation tracking of repositories. The interview also touches on the issues with package managers and the quantification of software vulnerability risks. Overall, the goal is to improve the efficiency and effectiveness of software component updates and leverage data to enhance security in enterprise and production environments.Learn more from The New Stack:Creating a 'Minimum Elements' SBOM Document in 5 MinutesEnhance Your SBOM Success with SLSA

Apache Airflow is an open-source platform for building machine learning pipelines. It allows users to author, schedule, and monitor workflows, making it well-suited for tasks such as data management, model training, and deployment. In a discussion on The New Stack Makers, three technologists from Amazon Web Services (AWS) highlighted the improvements and ease of use in Apache Airflow.Dennis Ferruzzi, a software developer at AWS, is working on updating Airflow's logging and metrics backend to the OpenTelemetry standard. This update will provide more granular metrics and better visibility into Airflow environments. Niko Oliveria, a senior software development engineer at AWS, focuses on reviewing and merging pull requests as a committer/maintainer for Apache Airflow. He has worked on making Airflow a more pluggable architecture through the implementation of AIP-51.Raphaël Vandon, also a senior software engineer at AWS, is contributing to performance improvements and leveraging async capabilities in AWS Operators, which enable seamless interactions with AWS. The simplicity of Airflow is attributed to its Python base and the operator ecosystem contributed by companies like AWS, Google, and Databricks. Operators are like building blocks, each designed for a specific task, and can be chained together to create workflows across different cloud providers.The latest version, Airflow 2.6, introduces sensors that wait for specific events and notifiers that act based on workflow success or failure. These additions aim to simplify the user experience. Overall, the growing community of contributors continues to enhance Apache Airflow, making it a popular choice for building machine learning pipelines.Check out the full article on The New Stack:How Apache Airflow Better Manages Machine Learning Pipelines

In this episode featuring Nima Negahban, CEO of Kinetica, the potential impact of generative AI tools like ChatGPT on businesses and organizations is discussed. Negahban highlights the transformative potential of generative AI when combined with data analytics. One use case he mentions is an "Alexa for all your data," where real-time queries can be made about store performance or product underperformance in specific weather conditions. This could provide organizations with a new level of visibility into their operations.Negahban identifies two major challenges in the generative AI space. The first is security, especially when using internal data to train AI models. The second challenge is ensuring accuracy in AI outputs to avoid misleading information. However, he emphasizes that generative AI tools, such as GitHub Copilot, can bring a new expectation of efficiency and innovation for developers.The future of generative AI in the enterprise involves discovering how to orchestrate these models effectively and leverage them with organizational data. Negahban mentions the growing interest in vector search and vector database capabilities to generate embeddings and perform embedding search. Kinetica's processing engine, coupled with OpenAI technology, aims to enable ad hoc querying against natural language without extensive data preparation, indexing, or engineering.Check out the episode to hear more about how the integration of generative AI and data analytics presents exciting opportunities for businesses and organizations, providing them with powerful insights and potential for creativity and innovation.Read more about Generative AI on The New StackIs Generative AI Augmenting Our Jobs, or About to Take Them?Generative AI: How to Choose the Optimal DatabaseHow Will Generative AI Change the Tech Job Market?Generative AI: How Companies Are Using and Scaling AI Models

In this episode of The New Stack Makers from KubeCon EU 2023, Rob Barnes, a senior developer advocate at HashiCorp, discusses how their networking service, Consul, allows users to incorporate containers or virtual machines into their workflows without imposing container usage. Consul, an early implementation of service mesh technology, offers a full-featured control plane with service discovery, configuration, and segmentation functionalities. It supports various environments, including traditional applications, VMs, containers, and orchestration engines like Nomad and Kubernetes.Barnes explains that Consul can dictate which services can communicate with each other based on rules. By leveraging these capabilities, HashiCorp aims to make users' lives easier and software more secure.Barnes emphasizes that there are misconceptions about service mesh, with some assuming it is exclusively tied to container usage. He clarifies that service mesh adoption should be flexible and meet users wherever they are in their technology stack. The future of service mesh lies in educating people about its role within the broader context and addressing any knowledge gaps.Join Rob Barnes and our host, Alex Williams, in exploring the evolving landscape of service mesh and understanding how it can enhance workflows.Find out more about HashiCorp or the biggest news from KubeCon on The New Stack:HashiCorp Vault Operator Manages Kubernetes SecretsHow HashiCorp Does Site Reliability EngineeringA Boring Kubernetes Release

What did software engineers at KubeCon say about how AI is coming up in their work? That's a question we posed Taylor Dolezal, head of ecosystem for the Cloud Native Computing Foundation at KubeCon in Amsterdam. Dolezal said AI did come up in conversation."I think that when it's come to this, typically with KubeCons, and other CNCF and LF events, there's always been one or two topics that have bubbled to the top," Dolezal said.At its core, AI surfaces a data issue for users that correlates to data sharing issues, said Dolezal in this latest episode of The New Stack Makers.Read more about AI and Kubernetes on The New Stack:3 Important AI/ML Tools You Can Deploy on KubernetesFlyte: An Open Source Orchestrator for ML/AI WorkflowsOvercoming the Kubernetes Skills Gap with ChatGPT Assistance

Kubernetes release 1.27 is boring, says Xander Grzywinski, a senior product manager at Microsoft.It's a stable release, Grzywinski said on this episode of The New Stack Makers from KubeCon Europe in Amsterdam."It's reached a level of stability at this point," said Grzywinski. "The core feature set has become more fleshed out and fully realized.The release has 60 total features, Grzywinski said. The features in 1.27 are solid refinements of features that have been around for a while. It's helping Kubernetes be as stable as it can be.Examples?It has a better developer experience, Grzywinski said. Storage primitives and APIs are more stable.

The mystery and miracle of flight sparked Ev Kontsevoy’s interest in engineering as a child growing up in the Soviet Union.“When I was a kid, when I saw like airplane flying over, I was having a really hard time not stopping and staring at it until it's gone,” said Kontsevoy, co-founder and CEO of Teleport, said in this episode of the Tech Founders Odyssey podcast series. “I really wanted to figure out how to make it fly.”Inevitably, he said, the engineering path led him to computers, where he was thrilled by the power he could wield through programming. “You're a teenager, no one really listens to you yet, but you tell a computer to go print number 10 ... and then you say, do it a million times. And the stupid computer just prints 10 million. You feel like a magician that just bends like machines to your will.”In this episode of the series, part of The New Stack Makers podcast, Kontsevoy discussed his journey to co-founding Teleport, an infrastructure access platform, with TNS co-hosts Colleen Coll and Heather Joslyn.

Developer tool integration and AI differentiate workflows to achieve that "fluid" state developers strive for in their work.Amazon CodeCatalyst and Amazon CodeWhisperer exemplify how developer workflows are accelerating and helping to create these fluid states. That's a big part of the story we hear from Harry Mower, director AWS DevOps Services, and Doug Seven, director, Software Development, AWS CodeWhisperer, from our recording in Seattle earlier in April for this week's AWS Developer Innovation Day.CodeCatalyst serves as an end-to-end integrated DevOps toolchain that provides developers with everything they need to go from planning through to deployment, Mower said. CodeWhisperer is an AI coding companion that generates whole-line and full-line function code recommendations in an integrated development environment (IDE).CodeWhisperer is part of the IDE, Seven said. The acceleration is two-fold. CodeCatalyst speeds the end-to-end integration process, and CodeWhisper accelerates writing code through generative AI.

Just as everyone was heading out to the New Year's holidays last year, CTO Rob Zuber got a surprise of a most unwelcome sort. A customer alerted CircleCI to suspicious GitHub OAuth activity. Although the scope of the attack appeared limited, there was still no telling if other customers of the DevOps-friendly continuous integration and continuous delivery platform were impacted.This notification kicked off a deeper review by CircleCI’s security team with GitHub, and they rotated all GitHub OAuth tokens on behalf of their customers. On January 4, the company also made the difficult but necessary decision to alert customers of this “security instance,” asking them to immediately rotate any and all stored secrets and review internal logs for any unauthorized access.In this latest episode of The New Stack Makers podcast, we discuss with Zuber the attack and how CircleCI responded. We also talk about what other companies should do to avoid the same situation, and what to do should it happen again.

Feature flags, the toggles in software development that allow you to turn certain features on or off for certain customers or audiences, offer release management at scale, according to Karishma Irani, head of product at LaunchDarkly.But they also help unleash innovation, as she told host Heather Joslyn of The New Stack in this episode of The New Stack Makers podcast. And that points the way to a future where the potential for easy testing can inspire new features and products, Irani said.“We've observed that when the risk of releasing something is lowered, when the risk of introducing bugs in production or breaking, something is reduced, is lowered, our customers feel organically motivated to be more innovative and think about new ideas and take risks,” she said.

Hoi Europe and beyond!Once again it is time for cloud native enthusiasts and professionals to converge and discuss cloud native computing in all its efficiency and complexity. The Cloud Native Computing Foundation's KubeCon+CloudNativeCon 2023 is being held later this month in Amsterdam, April 18 - 21, at the Rai Convention Centre.In this latest edition of The New Stack podcast, we spoke with two of the event's co-chairs who helped define this year's themes for the show, which is expected to draw over 9,000 attendees: Aparna Subramanian, Shopify's Director of Production Engineering for Infrastructure; and Cloud Native Infra and Security Enterprise Architect Frederick Kautz.

s the end of programming nigh?If you ask Matt Welsh, he'd say yes. As Richard McManus wrote on The New Stack, Welsh is a former professor of computer science at Harvard who spoke at a virtual meetup of the Chicago Association for Computing Machinery (ACM), explaining his thesis that ChatGPT and GitHub Copilot represent the beginning of the end of programming.Welsh joined us on The New Stack Makers to discuss his perspectives about the end of programming and answer questions about the future of computer science, distributed computing, and more.Welsh is now the founder of fixie.ai, a platform they are building to let companies develop applications on top of large language models to extend with different capabilities.For 40 to 50 years, programming language design has had one goal. Make it easier to write programs, Welsh said in the interview.Still, programming languages are complex, Welsh said. And no amount of work is going to make it simple.

Speed is a recurring theme in this episode of The Tech Founder Odyssey. Also, timing.Eilon Elhadad and Eylam Milner, who met while serving in the Israeli military, discovered that source code leak was a hazardous side effect of businesses’ need to move fast and break things in order to stay competitive.“Every new business challenge leads to a new technological solution,” said Elhadad in this episode of The New Stack's podcast series. “The business challenge was to deliver product faster to the business; the solution was to build off the supply chain. And then it leads to a new security attack surface.”Discovering this problem, and finding a solution to it, put Milner and Elhadad in the right place at the right time — just as the tech industry was beginning to rally itself to deal with this issue and give it a name: software supply chain security.It led them to co-found Argon Security, which was acquired by Aqua Security in late 2021, Elhadad told The New Stack, a year after Argon started.

Given the vulnerability of so many systems, it’s not surprising that cyberattacks on applications and APIs increased 82% in 2022 compared to the previous year, according to a report released this year by Imperva’s global threat researchers.What might rattle even the most experienced technologists is the sheer scale of those attacks. Digging into the data, Imperva, an application and data security company, found that the largest layer seven, distributed denial of service (DDoS) attack it mitigated during 2022 involved — you might want to sit down for this — more than 3.9 million API requests per second.“Most developers, when they think about their APIs, they’re usually dealing with traffic that’s maybe 1,000 requests per second, not too much more than that. Twenty thousand, for a larger API,” said Peter Klimek, director of technology at Imperva, in this episode of The New Stack Makers podcast. “So, to get to 3.9 million, it’s really staggering.”Klimek spoke to Heather Joslyn of TNS about the special challenges of APIs and cybersecurity and steps organizations can take to keep their APIs safe.The episode was sponsored by Imperva.

The 20th Annual Southern California Linux Expo (SCALE) runs Thursday through Sunday at the Pasadena Convention Center in Pasadena, Ca., featuring keynotes from notables such as Ken Thompson, the creator of Unix, said Ilan Rabinovich, one of the co-founders and conference chair for the conference on this week's edition of The New Stack Makers. "Honestly, most of the speakers we've had, you know, we got at SCALE in the early days, we just, we, we emailed them and said: 'Would you come to speak at the event?' We ran a call for proposals, and some of them came in as submissions, but a lot of it was just cold outreach. I don't know if that succeeded, because that's the state of where the community was at the time and there wasn't as much demand or just because or out of sheer dumb luck. I assure you, it wasn't skill or any sort of network that we like, we just, you know, we just we managed to, we managed to do that. And that's continued through today. When we do our call for papers, we get hundreds and hundreds of submissions, and that makes it really hard to choose from." Rethinking Web Application Firewalls Thompson, who turned 80 on February 4 (Happy Birthday, Mr. Thompson), created Unix at Bell Labs. He worked with people like Robert Griesemer and Rob Pike on developing the Go programming language and other projects over the years, including Plan 9, UTF-8, and more. Rabinovich is pretty humble about the keynote speakers that the conference attracts. He and the conference organizers scoured the Internet and found Thompson's email, who said he'd love to join them. That's how they attracted Lawrence Lessig, the creator of the Creative Commons license, who spoke at SCALE12x in 2014 about the legal sides of open source, content sharing, and free software. "I wish I could say, we have this very deep network of connections," Rabinovich said. "It's just, these folks are surprisingly approachable, despite, you know, even after years and years of doing amazing work." SCALE is the largest community-run open-source and free software conference in North America, with roots befitting an event that started with a group of college students wanting to share their learnings about Linux. Rabinovitch was one of those college students attending UCSB, the University of California, Santa Barbara. "A lot of the history of SCALE comes from the LA area back when open source was still relatively new and Linux was still fairly hard to get up and running," Rabinovitch said. "There were LUGS (Linux User Groups) on every corner. I think we had like 25 LUGS in the LA area at one point. And so so there was a vibrant open source community.' Los Angeles's freeways and traffic made it difficult to get the open source community together. So they started LUGFest. They held the day-long event at a Nortel building until the telco went belly up. So, as open source people tend to do, they decided to scale, so to speak, the community gatherings. And so SCALE came to be – led by students like Rabinovitch. The conference started with a healthy community of 200 to 250 people. By the pandemic, 3,500 people were attending. For more about SCALE, listen to the full episode of The New Stack Makers wherever you get your podcasts.

When she was a student in her native Israel, Shira Shamban was a self-proclaimed “geek.” But, unusually for a tech company founder and CEO, not a computer geek. Shamban was a science nerd, with her sights set on becoming a doctor. But first, she had to do her state-mandated military service. And that’s where her path diverged. In the military, she was not only immersed in computers but spent years working in intelligence; she stayed in the service for more than a decade, eventually rising to become head of an intelligence sector for the Israeli Defense Forces. At home, she began building her own projects to experiment with ideas that could help her team. “So that kind of helped me not to be intimidated by technology, to learn that I can learn anything I want by myself,” said Shamban, co-founder of Solvo, a company focused on data and cloud infrastructure security. “And the most important thing is to just try out things that you learn.” To date, Solvo has raised about $11 million through investors like Surround Ventures, Magenta Venture Partners, TLV Partners and others. In this episode of The New Stack Makers podcast series The Tech Founder Odyssey, Shamban talked to Heather Joslyn and Colleen Coll of TNS about her journey.In-Person TeamworkShamban opted to stay in the technology world, nurturing a desire to eventually start her own company. It was during a stint at Dome9, a cloud security company, that she met her future Solvo co-founder, David Hendri — and built a foundation for entrepreneurship. “After that episode, I got the guts,” she said. “Or I got stupid enough.” Hendri, now Solvo’s chief technology officer, struck Shamban as having the right sensibility to be a partner in a startup. At Dome9, she said, “very often, I used to stay up late in the office, and I would see him as well. So we'd grab something to eat.” Their casual conversations quickly revealed that Hendri was often staying late to troubleshoot issues that were not his or his team’s responsibility, but simply things that someone needed to fix. That sense of ownership, she realized, “is exactly the kind of approach one would need to bring to the table in a startup.” The mealtime chats that started Solvo have carried over into its current organizational culture. The company employs 20 people; workers based in Tel Aviv are expected to come to the office four days a week. Hendri and Shamban started their company in the auspicious month of March 2020, just as the Covid-19 pandemic started. While many companies have moved to all-remote work, Solvo never did. “We knew we wanted to sit together in the same room, because the conversations you have over a cup of coffee are not the same ones that you have on a chat, and on Slack,” the CEO said. “So that was our decision. And for a long time, it was an unpopular decision.” As the company scales, finding employees who align with its culture can make recruiting tricky, Shamban said. It's not only about your technical expertise, it's also about what kind of person you are,” she said. “Sometimes we found very professional people that we didn't think would make a good fit to the culture that we want to build. So we did not hire them. And in the boom times, when it was really hard to hire engineers. “These were tough decisions. But we had to make them because we knew that building a culture is easier in a way than fixing a culture. Listen to the full episode to hear more about Shamban's journey.

At Cloud Native Security Con, we sat down with Solo.io's Marino Wijay and Jim Barton, who discussed how service mesh technologies have matured, especially now with the removal of sidecars in Ambient Mesh that it developed with Google. Ambient Mesh is "a new proxy architecture that, according to the Solo.io site, "moves the proxy to the node level for mTLS and identity. It also allows a policy-enforcement policy to manage Layer 7 security filters and policies. A sidecar is a mini-proxy, a mini-firewall, like an all-in-one router, said Wijay, who does developer relations and advocacy for Solo. A sidecar receives instructions from an upstream control plane. "Now, one of the things that we started to realize with different workloads and different patterns of communication is that not all these workloads need a sidecar or can take advantage of the sidecar," Wijay said. "Some better operate without the sidecar." Ambient Mesh reflects the maturity of service mesh and the difference between day one and day two operations, said Barton, a field engineer with Solo. "Day one operations are a lot about understanding concepts, enabling developers, initial configurations, that sort of thing," Barton said. "The community is really much more focused and Ambient Mesh is a good example of this on day two concerns. How do I scale this? How do I make it perform in large environments? How can I expand this across clusters, clusters in multiple zones in multiple regions, that sort of thing? Those are the kinds of initiatives that we're really seeing come to the forefront at this point." With the maturity of service mesh comes the users. In the context of security, that means the developer security operations person, Barton said. It's not the developer's job to connect services. Their job is to build out the services. "It's up to the platform operator, or DevSecOps engineers to create that, that fundamental plane or foundation for where you can deploy your services, and then provide the security on top of it," Barton said. The engineers then have to configure it and think it through. "How do I know who's doing what and who's talking to who, so that I can start forming my zero trust posture?," Barton said.

Here's a breakdown of what we cover: Cloud IDEs will mature as GitHub's Codespaces platform gains acceptance through its integration into the GitHub service. Other factors include new startups in the space, such as GitPod, which offers a secure, cloud-based IDE, and Uptycs, which uses telemetry data to lock-down developer environments. "So I think you'll, you're just gonna see more people exposed to it, and they're gonna be like, 'holy crap, this makes my life a lot easier '." FinOps reflects the more stringent views on managing costs, focusing on the efficiency of resources that a company provides for developers. The focus also translates to the GreenOps movement with its emphasis on efficiency. Software bill of materials (SBOMs) will continue to mature with Sigstore as the project with the fastest expected adoption. Witness, from Telemetry Project, is another project. The SPDX community has been at the center of the movement for over a decade now before people cared about it. GitOps and Open Telemetry: This year, KubeCon submissions topics on GitOps were super high. OpenTelemetry is the second most popular project in the CNCF, behind Kubernetes. Platform engineering is hot. Anisczyk cites Backstage, a CNCF project, as one he is watching. It has a healthy plugin extension ecosystem and a corresponding large community. People make fun of Jenkins, but Jenkins is likely going to be around as long as Linux because of the plugin community. Backstage is going along that same route. WebAssembly: "You will probably see an uptick in edge cases, like smaller deployments as opposed to full-blown cloud-based workloads. Web Assembly will mix with containers and VMs. "It's just the way that software works." Kubernetes is part of today's distributed fabric. Linux is now everywhere. Kubernetes is going through the same evolution. Kubernetes is going into airplanes, cars, and fast-food restaurants. "People are going to focus on the layers up top, not necessarily like, the core Kubernetes project itself. It's going to be all the cool stuff built on top."

Everyone in the community was surprised by ChatGPT last year, which a web service responded to any and all user questions with a surprising fluidity. ChatGPT is a variant of the powerful GPT-3 large language model created by OpenAI, a company owned by Microsoft. It is still a demo though it is pretty clear that this type of generative AI will be rapidly commercialized. Indeed Microsoft is embedding the generative AI in its Bing Search service, and Google is building a rival offering. So what are smaller businesses to do to ensure their messages are heard to these machine learning giants? For this latest podcast from The New Stack, we discussed these issues with Ryan Johnston, chief marketing officer for Writer. Writer has enjoyed an early success in generative AI technologies. The company's service is dedicated to a single mission: making sure its customers' content adheres to the guidelines set in place. This can include features such as ensuring the language in the copy matches the company's own designated terminology, or making sure that a piece of content covers all the required topic points, or even that a press release has quotes that are not out of scope with the project mission itself. In short, the service promises "consistently on-brand content at scale," Johnston said. "It's not taking away my creativity. But it is doing a great job of figuring out how to create content for me at a faster pace, [content] that actually sounds like what I want it to sound like." For our conversation, we first delved into how the company was started, its value proposition ("what is it used for?") and what role that AI plays in the company's offering. We also delve a bit into the technology stack Writer deploys to offer these services, as well as what material the Writer may require from their customers themselves to make the service work. For the second part of our conversation, we turn our attention to how other companies (that are not search giants) can get their message across in the land of large language models, and maybe even find a few new sources of AI-generated value along the way. And, for those public-facing businesses dealing with Google and Bing, we chat about how they should they refine their own search engine optimization (SEO) strategies to be best represented in these large models? One point to consider: While AI can generate a lot of pretty convincing text, you still need a human in the loop to oversee the results, Johnston advised. "We are augmenting content teams copywriters to do what they do best, just even better. So we're scaling the mundane parts of the process that you may not love. We are helping you get a first draft on paper when you've got writer's block," Johnston said. "But at the end of the day, our belief is there needs to be a great writer in the driver's seat. [You] should never just be fully reliant on AI to produce things that you're going to immediately take to market."