Manufacturing Hub

Digital twins and machine learning are redefining batch optimization in manufacturing. Learn how centerlining models can catch quality issues in real time before they become irreversible.Concepts like digital twins, golden batch profiles, and statistical process control have long promised more than they delivered. Virag Vora of Twin Thread argues that layering machine learning on top of these ideas is what finally brings them to life. In this context, a digital twin is entirely data centric: a real time and historical representation of a process that serves as the foundation for AI models.The core use case is batch centerlining. The model compares current conditions against historically successful profiles, segmented by raw material source, product type, and seasonality. An orange juice manufacturer uses Twin Thread to determine whether incoming fruit should be sold fresh or routed to concentrate based on seasonal sugar content. The model identifies contributing variables in real time and alerts operators before a batch drifts beyond recovery.Twin Thread tackles the "not enough data" objection head on. With over 60 connectors, the platform works with the fragmented data reality of most manufacturing sites. Even low frequency data can train a useful model that quantifies what higher resolution instrumentation would unlock.Virag draws a clear line between ML and LLMs for process control. ML models trained on historical data produce deterministic outputs trusted for real time guidance on machine settings. LLMs excel at document retrieval and natural language interaction but are not suited for recommending set points on a live line. Twin Thread layers both: ML handles optimization, while Twin Thread Advisor lets users interrogate data and configure models through conversation.The standout proof point is Hills Pet Nutrition. After three years on Twin Thread, their models automatically feed recommendations into live production. That closed loop followed a deliberate path from human validation to A/B trials to automated execution with operator opt out.About Virag VoraVirag Vora is a solutions professional at Twin Thread, a platform that combines data centric digital twins with machine learning to optimize manufacturing processes. With a background in chemical engineering, Virag began his career deploying MES and DCS systems in biotech and pharma before joining Tulip and then Twin Thread. He helps manufacturers connect their existing data infrastructure to AI powered optimization across batch, continuous, and hybrid processes.Timestamps0:00 Introduction1:20 Virag's background in chemical engineering and industrial software6:30 Moving up the ISA 95 stack from DCS to MES and applications9:00 How AI reinvents digital twin, golden batch, and SPC concepts12:20 What a data centric digital twin actually looks like21:40 Where digital twins deliver the most value in manufacturing27:00 Seasonality, segmentation, and model training strategies36:00 Data prerequisites for deploying industrial AI41:40 Flavors of AI in manufacturing: ML, LLMs, and agentic workflows50:40 Closed loop AI control at Hills Pet Nutrition53:10 Personal project: Family Graph using knowledge graphs56:20 Prediction: operators as human digital twinsReferencesTwin Thread: https://twinthread.comThis episode is sponsored byMaintainX is an AI powered maintenance and operations platform that helps technicians get the answers they need instantly so they can focus on getting assets back online. Learn more about how MaintainX supports frontline manufacturing teams.https://maintainx.comAbout Your HostsVladimir Romanov is a co-host of The Manufacturing Hub Podcast and the founder of Joltek, an independent manufacturing and industrial automation consulting firm specializing in modernization strategy, digital transformation, and workforce development. Joltek works with manufacturers and investors to de-risk modernization and build the internal capability to sustain results.Connect with Vlad: https://www.linkedin.com/in/vladimirromanov/Want to go deeper? Vlad and the team at Joltek have covered related topics here:Edge Computing, AI, and the Value of Manufacturing Data: https://www.joltek.com/blog/edge-computing-ai-value-manufacturing-dataDigital Transformation in Manufacturing: https://www.joltek.com/blog/digital-transformation-in-manufacturingDave Griffith is a co-host of The Manufacturing Hub Podcast and founder of Capelin Solutions, an industrial automation firm helping manufacturers adopt smart manufacturing technology. He brings 15 years of experience in industrial automation and digital transformation.Connect with Dave: https://www.linkedin.com/in/davegriffith23/Subscribe to Manufacturing Hub: https://www.manufacturinghub.liveLinkedIn: https://www.linkedin.com/company/manufacturing-hub-networkYouTube: https://www.youtube.com/@ManufacturingHub

Physical AI is arriving on factory floors ahead of schedule, and Vention is already deploying it on applications four automation integrators failed to crack.François Giguère, CTO of Vention, draws a precise line between agentic AI and physical AI. Agentic systems process data and return data. Physical AI controls motion and actuation that produce real world consequences on a factory floor where a hundred percent uptime is the only acceptable standard. Giguère has spent a decade helping build Vention, a platform that lets manufacturers design robotic cells in 3D, program them through natural language, simulate them in a browser, and receive the physical machine shipped in modular components like an industrial kit. With a team of 95 engineers and three years as CTO, he brings a grounded perspective on where AI delivers real value in industrial automation and where it still falls short.The design, automate, simulate workflow at Vention represents one of the most complete implementations of AI-powered machine engineering currently in production. In the design phase, customers build systems from a modular component library. In the automate phase, an AI agent converts natural language prompts into Python control code for the entire cell including robot arms, conveyors, vision systems, and grippers. The program is validated in simulation before a single component ships. This is made possible by Vention's motion streaming architecture: instead of treating the robot as the master controller the way KUKA KRL does, Vention brings all motion planning, inverse kinematics, forward kinematics, blending, and trajectory optimization into its own software stack. The robot becomes a passive component consuming a motion stream, and the entire machine becomes programmable from a single unified codebase that AI tools excel at generating. Giguère notes that Vention's choice to use Python as the programming language for automation control gives their AI tools a measurable edge over environments built on structured text or ladder logic.Vention's two physical AI products are GRIP (Generalized Robotics Intelligence Pipeline) and Rapid AI Operator, a modular bin picking application built on top of GRIP. The technology relies on transformer-based foundation models.About François GiguèreFrançois Giguère is the CTO of Vention, an industrial automation platform where manufacturers design, program, simulate, and deploy robotic systems entirely online. Employee number four at the company, he has contributed to Vention's growth for over 10 years and leads a team of 95 engineers. He holds a background in electrical engineering and real-time embedded software development.Learn more: https://vention.ioTimestamps0:00 Introduction and welcome1:00 François Giguère's background and Vention overview2:20 How AI spans Vention's internal tools and customer products4:00 Why embedded and robotics code is harder for AI to generate7:00 Design, automate, simulate: Vention's three-stage AI workflow13:50 Motion streaming: one unified controller for all robot brands18:20 Defining physical AI versus agentic AI20:10 GRIP pipeline and Rapid AI Operator22:40 Case study: MacAlpine Plumbing bin picking with foundation models39:40 Nvidia GTC impressions: agentic AI eclipsing physical AI46:20 Edge versus cloud: why real-time inference stays on-prem56:10 Predictions: physical AI roadmap and the VLA timelineThis episode is sponsored by:MaintainX helps maintenance and operations teams work smarter by putting critical information directly in the hands of technicians. According to MaintainX, technicians spend up to 40 percent of their time searching for answers and responding to radio calls rather than fixing assets.https://www.maintainx.comAbout Your HostsVladimir Romanov is a co-host of The Manufacturing Hub Podcast and the founder of Joltek, an independent manufacturing and industrial automation consulting firm specializing in modernization strategy, digital transformation, and workforce development.Connect with Vlad: https://www.linkedin.com/in/vladromanov/Want to go deeper? Vlad and the team at Joltek have covered related topics here:Industrial Robotics: https://www.joltek.com/blog/industrial-roboticsEdge Computing and AI Value in Manufacturing Data: https://www.joltek.com/blog/edge-computing-ai-value-manufacturing-dataDave Griffith is a co-host of The Manufacturing Hub Podcast and founder of Capelin Solutions, an industrial automation firm helping manufacturers adopt smart manufacturing technology. He brings 15 years of experience in industrial automation and digital transformation.Connect with Dave: https://www.linkedin.com/in/davegriffith23/Subscribe to Manufacturing Hub: https://www.manufacturinghub.liveLinkedIn: https://www.linkedin.com/company/manufacturing-hub-networkYouTube: https://www.youtube.com/@ManufacturingHub

AI in manufacturing is no longer a strategy reserved for the boardroom. It is a tool for the technician on the plant floor, and the results are already showing up in real operations worldwide.Most digital transformation strategies in manufacturing are built for desk workers on the carpeted side of the building, not the operators and technicians keeping production running on the concrete floor. AI platforms have historically been designed for white collar knowledge workers with time to navigate complex systems, leaving the frontline worker as an afterthought. Nick Haase recognized this gap when building MaintainX in 2018, and it became the foundational design principle behind everything the company built. The result is a platform now serving nearly 14,000 customers across manufacturing, food and beverage, facilities management, and any industry that depends on physical assets staying operational.The core thesis Nick brings to this conversation is that the person with no purchasing authority and no budget is the single most important factor in whether a digital transformation project succeeds or fails. That person is the frontline technician. Building for that user first required a mobile experience so intuitive that no training was needed, one that met workers in the flow of existing work rather than pulling them out of it. If your team needs a 300 page manual to use the platform, the adoption battle is already lost.The skilled labor shortage in manufacturing is not a forecast. The United States is projected to have more than 3 million manufacturing jobs unfilled by 2030, driven largely by retirement of experienced workers who have spent decades building institutional knowledge. That knowledge cannot be transferred through a job posting. MaintainX attacks this through AI powered voice note capture at work order closeout. Technicians leave a verbal description of what they found and fixed. The platform transcribes it across any language or accent, standardizes it, and builds a living knowledge base that outlasts the retirements of the people who created it. For organizations with similar equipment across dozens of sites, that knowledge becomes portable across locations and years.About Nick HaaseNick Haase is a co-founder of MaintainX, a frontline work execution platform for maintenance, reliability, SOPs, safety, and compliance serving nearly 14,000 customers across manufacturing and other asset-intensive industries. Nick is also the host of The Wrench Factor podcast.Connect with Nick: https://www.linkedin.com/in/nickhaase/Timestamps0:00 Introduction1:30 Nick Haase and MaintainX Background7:20 Where AI Fits for Frontline Workers10:00 What Data Foundations Are Needed for AI13:30 Why Frontline Adoption Determines Digital Transformation Success16:40 The Skilled Labor Shortage and Retirement Wave18:30 Voice Notes and AI Powered Knowledge Capture25:30 Overcoming Change Management and AI Skepticism34:50 Guardrails and Safe AI for Industrial Environments45:10 Embedding AI in the Flow of Work48:30 AI Agents for Parts Forecasting and Automation55:50 Predict the Future: Maintenance as a Growth CenterReferencesMaintainX: https://www.maintainx.comThe Wrench Factor Podcast: https://podcasts.apple.com/us/podcast/the-wrench-factor/id1809000028Origins of Efficiency by Brian Potter: https://www.amazon.com/dp/B0FJG6ZKKJInductive Automation Ignition: https://inductiveautomation.comThis episode is sponsored by MaintainXTechnicians spend up to 40 percent of their time looking for answers rather than fixing equipment. MaintainX puts AI powered knowledge tools directly in the flow of work so frontline teams get the right information in seconds.https://www.maintainx.comAbout Your HostsVladimir Romanov is a co-host of The Manufacturing Hub Podcast and the founder of Joltek, an independent manufacturing and industrial automation consulting firm specializing in modernization strategy, digital transformation, and workforce development. Joltek works with manufacturers and investors to de-risk modernization and build the internal capability to sustain results.Connect with Vlad: https://www.linkedin.com/in/vladimirromanov/Joltek: https://www.joltek.com/blog/digital-transformation-in-manufacturingJoltek: https://www.joltek.com/blog/root-causes-downtime-industrial-automationDave Griffith is a co-host of The Manufacturing Hub Podcast and founder of Capelin Solutions, an industrial automation firm helping manufacturers adopt smart manufacturing technology. He brings 15 years of experience in industrial automation and digital transformation.Connect with Dave: https://www.linkedin.com/in/davegriffith23/Subscribe to Manufacturing Hub: https://www.manufacturinghub.liveLinkedIn: https://www.linkedin.com/company/manufacturing-hub-networkYouTube: https://www.youtube.com/@ManufacturingHub

Industrial AI is getting a lot of attention in manufacturing right now, but one of the biggest questions is still the most practical one. How do you turn plant data, process knowledge, and operational constraints into something that actually creates value? In this episode of Manufacturing Hub, Vlad Romanov and Dave Griffith sit down with Konstantin Paradizov of Eukodyne for a detailed conversation on what industrial AI looks like when it is applied by people who understand manufacturing, MES, process improvement, data architecture, and the realities of the plant floor.What makes this discussion especially valuable is that it does not stay at the surface level. Konstantin shares how his background moved from pharma into food and beverage, how Lean Six Sigma and process thinking shaped his approach, and why many of the best opportunities in manufacturing still begin with understanding the actual workflow before talking about software. The conversation explores a theme that comes up again and again in industrial transformation: the biggest gains often do not come from adding more technology first. They come from understanding the problem clearly, identifying what information matters, validating assumptions with the people doing the work, and then using the right mix of tools to move faster.A major part of this episode focuses on the real use of AI in consulting and discovery. Konstantin explains how his team uses secure transcription workflows, on premises AI infrastructure, cloud models, masking of sensitive information, iterative validation, and ROI driven reporting to create high value outputs in a fraction of the time that would have been required even a year or two ago. This is an important point for manufacturers, system integrators, software teams, and plant leaders. AI is not just something that sits in front of an operator as a chatbot. It can be used behind the scenes to accelerate analysis, strengthen recommendations, shorten discovery, improve documentation, and reduce the cost of getting to a better answer.The technical section of this episode is especially strong for anyone working in industrial automation, OT data systems, or applied AI. The discussion covers on premises compute, Nvidia based edge hardware, Linux environments, Docker containers, RAG workflows, vector databases, knowledge graphs, MQTT pipelines, HiveMQ, Mosquitto, n8n, Claude Code, Cursor, Gemini, OpenRouter, and the tradeoffs between frontier models in the cloud and smaller or open models deployed closer to the process. One of the clearest takeaways is that manufacturers should not start with the biggest model or the most exciting headline. They should start with the problem, the constraints, the data path, and the economics of the solution.Vlad also pushes on an issue that matters to almost every manufacturer trying to prepare for AI. If you collect massive amounts of plant data into historians, cloud platforms, and enterprise systems, is that enough to create value later? Konstantin’s answer is thoughtful and realistic. More data alone does not automatically lead to better outcomes. You still need filtering, context, prioritization, architecture, and a disciplined way to separate signal from noise.Learn more about Joltek here:https://www.joltek.com/serviceshttps://www.joltek.com/services/service-details-it-ot-architecture-integrationConnect with our guest:Konstantin Paradizovhttps://www.linkedin.com/in/konstantin-paradizov/Learn more about Eukodyne:https://eukodyne.com/Follow Manufacturing Hub for more conversations on industrial AI, digital transformation, OT architecture, SCADA, MES, industrial data strategy, systems integration, and the future of manufacturing technology.Timestamps00:00 Welcome and introduction to industrial AI applications01:50 Konstantin’s background from pharma to manufacturing05:30 Why food and beverage offered major process improvement opportunities08:10 How to identify the right manufacturing opportunities to pursue13:10 Using AI to accelerate discovery, documentation, and customer value21:20 The on premises AI hardware stack and model selection strategy30:10 Why iterative validation still matters more than a first AI answer39:00 Claude Code, developer workflows, and practical AI tool stacks48:20 On premises versus cloud AI and how to think about the tradeoff54:10 Small models, low cost hardware, and edge deployment realities01:05:00 Plant data, historians, filtering, and separating signal from noise01:14:50 Predictions for industrial AI, career advice, and final recommendationsReferences and resources mentioned in the episodeMaintainXhttps://www.maintainx.com/Solve for Happyhttps://www.mogawdat.com/booksGeorge Orwell 1984https://www.penguinrandomhouse.com/books/326569/1984-by-george-orwell/George Orwell Animal Farmhttps://www.penguinrandomhouse.com/books/561805/animal-farm-by-george-orwell/

Manufacturing Hub is back with Episode 252, where co hosts Vlad Romanov and Dave Griffith break down what an AI survival guide should actually look like for manufacturing and industrial automation professionals. This is not a hype conversation about replacing people with magic software. It is a grounded discussion about what AI tools can do today, where they fail, why context and data quality matter so much, and how industrial teams should think about experimentation without losing sight of real operating constraints.In this episode, Vlad and Dave unpack the evolution many engineers and technical leaders have already felt in real time, from early prompt engineering, to agent based workflows, to MCP servers, skills, context management, and the growing cost of tokens and infrastructure. The conversation moves beyond generic AI commentary and into the reality of plant floor environments, where success depends on process knowledge, data architecture, OT constraints, cybersecurity, governance, and clear business value. One of the strongest themes throughout the episode is that manufacturers cannot skip the hard work of structuring data, understanding workflows, and defining use cases simply because AI tools are moving quickly.Vlad brings a very practical industrial lens to the discussion. Drawing on years of hands on experience across controls, manufacturing systems, plant modernization, and digital transformation, he explains why industrial AI has to start with operational context. A maintenance team, an engineering team, and a quality team do not need the same data, do not ask the same questions, and should not be handed the same AI workflows. That distinction matters. This conversation also highlights why the best industrial AI implementations will likely come from teams that combine domain expertise with strong technical execution, rather than generic AI shops trying to force a solution into environments they do not fully understand.Dave adds an important systems and adoption perspective, especially around cost, scaling, management expectations, and the danger of trying to prompt your way past foundational architecture work. Together, Vlad and Dave explore why manufacturers are interested in AI, why many are afraid of being left behind, and why so many projects still stall once they hit the realities of obsolete equipment, weak data models, fragmented systems, and unclear ownership of information. They also discuss deterministic logic versus LLM behavior, reporting workflows, industrial dashboards, PLC code generation concerns, and the practical question every manufacturer should ask before investing: what problem are we solving, for whom, and what is the measurable return?For those new to Vlad, he is an electrical engineer and manufacturing leader with deep experience across industrial automation, controls, data systems, OT architecture, modernization strategy, and plant operations. Through Joltek, Vlad works with manufacturers on digital transformation, IT OT architecture and integration, modernization planning, operational improvement, and technical workforce enablement. Learn more here:Joltek: https://www.joltek.com IT OT Architecture and Integration: https://www.joltek.com/services/service-details-it-ot-architecture-integrationIf you are a plant leader, controls engineer, systems integrator, OT architect, SCADA or MES practitioner, or simply someone trying to separate useful AI workflows from noise, this episode will give you a much more realistic framework for thinking about industrial AI adoption.Timestamps00:00 Welcome back and why this episode matters01:00 Setting up the industrial AI theme for the coming weeks03:10 From prompt engineering to structured AI workflows05:30 AI agents, parallel workflows, tokens, and context windows09:00 MCP tools, Playwright, and what new integrations unlock16:20 How Vlad researches AI and where useful information actually lives22:00 Real manufacturing problems versus AI in search of a problem29:40 Why industrial data architecture is harder than most people think37:00 OT expertise, workforce enablement, and who should build solutions45:40 Practical advice for manufacturers starting the AI journey50:30 Data governance, hallucinations, infrastructure, and cybersecurity57:20 What looks promising today in reporting, dashboards, and industrial applications

In this episode of Manufacturing Hub, Vlad and Dave sit down with Travis Cox and Kevin McCluskey from Inductive Automation to unpack what was actually proven at ProveIt and why it matters for teams trying to modernize plants without building a fragile mess of point to point integrations. If you have ever looked at a shiny demo and wondered what the real architecture looks like, how it scales beyond a single line, and what it takes to roll out across multiple sites without turning every change into a high risk event, this conversation is for you.Travis and Kevin walk through their ProveIt Enterprise B build and the thinking behind it. The core idea is simple but powerful: treat the factory like a system that needs a shared digital infrastructure, built on open standards, where data is contextualized and reusable. They break down how they used Ignition Edge close to PLCs for resiliency, local HMIs, and disciplined data modeling, then moved data through MQTT into a Unified Namespace so multiple applications can consume the same trusted signals and context. This is the difference between “we can connect to anything” and “we can scale without rewriting everything every time the business changes.” Open standards show up repeatedly in the conversation because ProveIt is specifically designed to force interoperability and practical implementation tradeoffs. Inductive Automation has also written about ProveIt as a place where MQTT, OPC UA, and SQL show up as real foundations rather than slogans.From there, the episode gets into the part that should make both OT and IT teams pay attention: modern deployment practices applied to industrial applications. Kevin outlines a clear maturity path from a single designer workflow to version control, then to containerized deployments, and finally to full GitOps style promotion across dev, staging, and production using tools like Argo CD, Helm, Kubernetes, and release promotion concepts that look like what the software world has used for years. Argo CD is explicitly built around Git repositories as the source of truth for desired state, which is exactly why it fits this style of deployment. The live portion of the conversation demonstrates how fast this can get when the infrastructure is treated as code: they spin up a brand new “site four” by submitting a form, generating a pull request, merging it, and letting the pipeline do the rest.Timestamps00:00 Welcome back and why this ProveIt recap matters01:35 Meet Travis Cox and Kevin McCluskey from Inductive Automation03:10 What ProveIt is and the key vendor questions it forces05:20 Enterprise B architecture overview from PLC to Edge to site to enterprise07:30 HMI walkthrough across liquid processing, filling, packaging, palletizing09:05 Why deploy Ignition Edge instead of only a centralized site gateway12:05 Design once, reuse everywhere and what that means for scaling quickly14:35 On prem realities versus cloud infrastructure in the ProveIt environment17:10 MCP, n8n workflows, and bringing live operational context into AI20:40 i3X style API access to models, history, and alarms for interoperability23:15 GitHub, Docker Compose, Helm, Kubernetes, Argo CD, Cargo and GitOps promotion36:55 Spinning up a new site live and what it changes for multi site rolloutsAbout the hostsVlad Romanov is an electrical engineer and MBA who has spent over a decade building and modernizing manufacturing systems across industrial automation, controls, and plant operations. Through Joltek, Vlad works with manufacturers to assess current state OT foundations, reduce modernization risk, improve reliability, and build internal capability through practical training and standards that stick.Dave Griffith co hosts Manufacturing Hub and brings a practitioner lens focused on what works on the plant floor, how architectures survive real constraints, and how industrial teams can modernize without breaking production.About the guestsTravis Cox is Chief Technology Evangelist at Inductive Automation and has spent over two decades helping customers and partners design scalable architectures, apply best practices, and deliver real solutions with Ignition.Kevin McCluskey is Chief Technology Architect at Inductive Automation and works with organizations on architecture decisions, platform direction, and enabling the next generation of industrial applications.Learn more about Joltekhttps://www.joltek.com/serviceshttps://www.joltek.com/book-a-modernization-consultation

This episode wraps up our Technology Modernization theme with a Siemens perspective that feels very grounded in what factories are actually dealing with right now. Brian Albrecht and Louis Hughes from the Siemens XD team walk through what they are seeing in the field across brownfield and greenfield conversations, why executives keep asking for industrial AI before the foundations are ready, and what it really takes to turn messy plant data into something you can trust for analytics, operations, and eventually AI enabled workflows.A big thread in this conversation is that modern manufacturing is not blocked by ambition, it is blocked by readiness. Everyone wants faster decisions, fewer surprises, and higher uptime, but the path there usually starts with boring work that is not optional. Data transparency across machine, plant, MES, and cloud layers. A clear definition of what real time actually needs to mean for a given use case. And a plan to contextualize and orchestrate data so that AI does not get fed junk inputs. Brian and Louis explain how they approach those early customer conversations, how workshops turn vision into prioritized use cases, and why trust, pilots, and repeatability matter more than flashy demos when you are working in regulated or high consequence environments.If you have been hearing nonstop AI buzz but you are still wrestling with legacy controls, inconsistent tags, documentation that no one can find, and seven layers of security constraints, this episode is for you. We get into practical use cases like AI vision and anomaly detection, LLMs for tribal knowledge and troubleshooting workflows, and the idea of fast versus slow AI, meaning AI that must act during production versus AI that can analyze after the fact.Timestamps00:00 Welcome and why this episode closes the modernization theme02:10 Meet Brian Albrecht and Louis Hughes from the Siemens XD team05:25 Vertical differences across oil and gas, discrete, and process manufacturing07:50 What executives ask for right now beyond AI, factory of the future and data transparency10:50 Brownfield reality and why most modernization work starts with legacy systems12:30 The AI conversation when foundations are missing, meeting customers where they are15:10 Current AI use cases in manufacturing, downtime, throughput, LLMs, and vision18:10 What it means to be AI ready, data silos, contextualization, and orchestration23:50 Fast versus slow AI and why production time decisions are different from analytics25:30 Edge versus cloud architecture, latency, and where the data should live33:40 Cybersecurity, trust, and why perception can lag behind the technology36:50 Hallucinations, guardrails, and why recommendations usually come before automation51:10 Book recommendations, career advice, and future predictions for industrial AIAbout the hostsVlad Romanov is an electrical engineer with an MBA from McGill University and over a decade of experience in manufacturing and industrial automation. He has worked across large scale environments including Procter and Gamble, Kraft Heinz, and Post Holdings, and he now leads Joltek, helping manufacturers modernize systems, improve reliability, strengthen IT and OT architecture, and upskill technical teams through practical training and on site enablement.Dave Griffith is the cohost of Manufacturing Hub and an industrial automation practitioner who focuses on how modern technologies translate into real factory outcomes, from controls and data foundations to scalable implementation strategies.About the guestsBrian Albrecht started in electrical engineering and spent about a decade in systems integration in Oklahoma City focused on oil and gas, building SCADA, networking, and automation solutions and leading teams delivering real world projects. He now works with Siemens customers on building relationships and delivering solutions that create measurable value.Louis Hughes has roughly 20 years of manufacturing experience, starting in software development for manufacturing and engineering applications, then moving into solution architecture, services delivery, and experience center leadership. He now leads a smart manufacturing team, bringing a software and systems view into automation conversations focused on solving customer problems, not just deploying tools.Joltek Services - https://www.joltek.com/servicesContact Joltek - https://www.joltek.com/contactReferenced in the episodeProveIt Conference - https://www.proveitconference.com/Siemens - https://www.siemens.com/Crossing the Chasm by Geoffrey A Moorehttps://en.wikipedia.org/wiki/Crossing_the_ChasmExtreme Ownership by Jocko Willink and Leif Babinhttps://en.wikipedia.org/wiki/Extreme_Ownership

In this special ProveIt edition of Manufacturing Hub, Vlad Romanoff and Dave Griffith sit down with Amy Williams from Skellig Automation to unpack Enterprise C, a life sciences virtual factory built to look and feel like the reality inside many regulated facilities today. If you work around batch processes, compliance, historian projects, electronic batch records, or industrial data architecture, this conversation is a practical walkthrough of what it actually takes to turn raw signals into a story you can defend, improve, and scale.Amy has spent years working exclusively in life sciences manufacturing, starting deep in DeltaV automation for batch pharma and moving into digital transformation projects that focus on open architectures, modern data pipelines, and real operational outcomes. In this episode, she explains what Enterprise C is simulating, why it was designed as an Industry 3.0 style biotech startup, and what kind of data and documentation a vendor would have to wrestle with in the real world. The factory is producing a fictional enzyme using a fed batch fermentation process, and the UNS publishes realistic one second resolution batch data across four pieces of single use equipment including a mixer, a bioreactor, a chromatography skid, and a TFF skid.One of the most valuable parts of this episode is the reminder that data sitting in an MQTT broker is not inherently valuable. The value comes when the data is contextualized enough that different teams can use it without tribal knowledge, and when the resulting traceability helps you answer the questions that matter in life sciences. What happened during the batch, what changed compared to previous runs, what went out of spec, what documentation proves compliance, and what you should do next time to avoid losing a batch that can cost millions. Amy also explains why Enterprise C intentionally includes uncontextualized tags and paper files, because that is exactly where many facilities still are. The hard part is not connecting a sensor, the hard part is governance, agreement, and building a model that humans actually follow.You will also hear the crew dig into Smart Manufacturing Profiles and why standardizing information models is one of the clearest paths toward true interoperability. If you are tired of every site, every integrator, and every project reinventing the same pump, valve, and equipment model from scratch, this is the kind of conversation that helps frame why that problem keeps repeating and what might finally reduce it. The ProveIt format forces the questions that most conferences avoid, including what problem was solved, how it was done, how long it took, and what it cost. That is exactly why this conference has become a magnet for practitioners who care about the difference between a demo and a deployable solution.About the hostsVlad Romanoff is an industrial automation and manufacturing systems expert and the founder of Joltek. He has over a decade of experience modernizing control systems, data infrastructure, and plant operations across regulated and high throughput manufacturing environments.Dave Griffith is the cohost of Manufacturing Hub and a long time practitioner in industrial automation and manufacturing technology, focused on practical deployment and what actually works on the plant floor.About the guestAmy Williams works with Skellig Automation and has spent years in life sciences manufacturing, from DeltaV batch automation to digital transformation initiatives that focus on open architectures, data contextualization, and scalable modernization strategies.Timestamps 00:00 ProveIt edition intro and why this month is technology modernization 01:40 Who is Amy Williams and why Enterprise C matters this year 02:10 Amy’s background in life sciences, DeltaV, and digital transformation 03:30 Unified Namespace explained in plain language for life sciences 05:10 What Enterprise C publishes and what you will see in the MQTT broker 07:10 Why UNS in life sciences is about use cases, not buzzwords 10:10 Smart Manufacturing Profiles and reducing data model reinvention 11:10 What outcomes to expect including compliance and golden batch analysis 12:10 Enterprise C process overview from mixer to bioreactor to downstream 14:10 Bioreactor instrumentation and what operators still do manually 19:40 Why Enterprise C data is intentionally not contextualized 22:10 The real work of mapping signals to compliance stories and governance 25:10 What SM Profiles enable and why schema matters before data arrives 31:30 Why cost and time questions change everything at ProveIt 36:10 Cell counter files, batch records, and paper driven reality in many sites 45:10 What life sciences attendees should ask during Q and A 58:30 Vendors the team is excited to see and why non traditional players matter 01:02:20 Where to find Skellig at the conference and what they are bringingReferences and links mentioned Skellig Automation https://www.skellig.com/ProveIt Conferenc

In this episode of Manufacturing Hub, Vlad Romanov and Dave Griffith sit down with David for a practical, operator grounded conversation about industrial data, modernization, and what it actually takes to turn plant floor signals into business decisions. David has spent more than two decades in manufacturing across automotive, solar, and electric vehicles, and his story is a familiar one for a lot of us. He walked into a plant thinking he was there for a project, discovered PLCs in real time, and never left the factory world. From early days wiring up a SQL Server to pull line data instead of sending people out with stopwatches, to leading data and analytics and shaping MES and reporting strategy, this conversation stays focused on the messy middle where most factories live.A big theme here is that collecting data is not the same thing as creating information. As tooling has improved, connectivity, historians, SCADA, cloud storage, MQTT, and the modern ecosystem have made it easier to get signals out of machines. The hard part is deciding what matters, aligning stakeholders, and creating context that survives across teams and projects. David breaks down how real progress often starts with simple visibility, what is ruining your day, what is the biggest safety risk, what is the recurring quality miss, what is the downtime story you do not trust, then builds from there using workshops and iterative delivery instead of giant multi year “boil the ocean” programs.We also get into Unified Namespace, why it resonates with people who have been burned by tightly coupled ISA style integrations, and why change management is the hidden cost. If you are exploring UNS, this episode highlights the difference between drawing the box on a whiteboard and getting a whole organization to actually adopt consistent naming, context, and ownership. Then we finish with a grounded take on industrial AI. No hype, no doom. Just a realistic view of where AI helps today, where it breaks, and why context windows, documentation quality, and domain expertise still decide whether results are useful or dangerous.Timestamps00:00:00 Welcome and the month theme on technology modernization00:02:10 David’s background from automotive and the Tesla Fremont NUMMI era to data leadership00:05:10 The moment data became “real” and why proactive visibility drives safety and outcomes00:07:10 How Kaizen and Toyota Production System style problem solving creates demand for data00:11:50 Why modern tooling makes collection easier and why budget and commitment still decide success00:16:10 Starting points that work in the real world and the simplest visibility model that scales00:18:20 Unified Namespace explained through decoupling, context, and why the first attempt often fails00:23:50 Who really uses the data, operators, quality, engineering, and the “next factory” teams00:29:10 Defining KPIs when nobody has answers and using workshops to force prioritization00:34:20 What rollouts actually take, machine states, data structures, controls changes, and iteration00:40:10 Industrial AI reality check, where it helps today and why it is not running your factory00:51:10 Predicting the next few years, consolidation, pricing, and better integration with agentsAbout the hostsVlad Romanov is an industrial automation and manufacturing leader with over a decade of plant floor experience across major manufacturers. He is the founder of Joltek, where he helps teams modernize operations through IT and OT architecture, integration, reliability focused execution, and practical upskilling that actually sticks. Joltek works with manufacturers who need real outcomes, not buzzwords, and the work spans controls, data, networking, and operational performance.Dave Griffith is the co host of Manufacturing Hub and works at the intersection of manufacturing operations, technology modernization, and practical delivery. He focuses on helping teams bridge the gap between “we want data” and “we can run this plant better next quarter.”About the guestDavid has 25 plus years of manufacturing experience spanning automotive, solar manufacturing, and EVs. He started in plant floor automation and conveyance projects, then moved deeper into industrial data, MES, and analytics leadership. His recent work includes leading data and analytics, defining KPI strategy, and building the layers required to turn raw plant signals into usable business information.Links from Joltekhttps://www.joltek.com/blog/mastering-unified-namespace-uns-a-guide-to-data-driven-manufacturing-transformationhttps://www.joltek.com/blog/ultimate-guide-mqtt-manufacturingSubscribe for more conversations on manufacturing modernization, industrial data architecture, MES realities, and what works on the plant floor when the budget, people, and legacy systems are all real.

In this episode of Manufacturing Hub, Vlad Romanov and Dave Griffith sit down with Benson Hougland from Opto 22 to get brutally practical about what is actually running on shop floors today, and what it takes to move from legacy automation to modern, data ready operations without breaking what already works. If you have ever walked into a plant and seen a mix of decades old controllers, manual processes, islands of automation, and a few shiny modern pockets of connectivity, this conversation will feel very familiar. Benson has spent roughly three decades at Opto 22 and he has seen the full spectrum, from brownfield realities where nothing can go down, to greenfield expansions where teams can finally design with data, security, and integration in mind.A major thread in this discussion is the gap between “the machine runs” and “the business can learn from the machine.” Benson lays out why so many facilities still operate in a world of siloed equipment with minimal visibility, and why digital transformation stalls when the goal is vague or driven by trend chasing. The most actionable insight is simple: start with a real problem, win small, build trust in the data, and only then scale. That approach is how you avoid proof of concept purgatory, and it is also how you get leadership buy in without overpromising. If you are looking at industrial AI, it becomes even more critical, because manufacturing cannot tolerate hallucinated answers. Benson explains why industrial AI starts with sanctity of data, meaning clean, contextualized, trustworthy signals that an organization can actually act on.You will also hear a grounded take on why hardware still matters in 2026. Not because everyone wants to rip and replace working PLCs, but because modern plants need layered edge strategies that can extract the right data, protect legacy assets, and integrate upward using open methods.About the guestBenson Hougland is a long time leader at Opto 22, a US based manufacturer of industrial controllers, edge devices, and IO. He focuses on customer and integrator feedback, product strategy, and the practical challenges teams face when modernizing systems while keeping operations running. Opto 22 is known for building and manufacturing in the United States and for leaning into open connectivity approaches that help reduce lock in and simplify integration.About the hostsVlad Romanov is an electrical engineer with an MBA from McGill University and over a decade of experience delivering automation and modernization work across high performing manufacturing environments. Through Joltek, Vlad supports manufacturers with plant floor assessments, controls and OT architecture, system modernization planning, integration execution, and technical upskilling so teams can own their systems long term. Vlad’s work consistently sits at the intersection of reliability, operational execution, and the realities of IT and OT convergence, with a focus on what is feasible in real facilities, not just what looks good in a slide deck.Dave Griffith is a long time manufacturing and automation practitioner focused on bridging the gap between modern technology conversations and what is practical on the plant floor. Dave brings a systems mindset to modernization, with a strong emphasis on outcomes, maintainability, and the human factors that decide whether projects scale or stall.If this episode resonates and you are navigating modernization decisions, especially around OT networking, data infrastructure, platform selection, or plant floor security, Joltek can help you evaluate your current state, define a realistic target architecture, and build a roadmap that your team can execute.Joltek linkshttps://www.joltek.com/serviceshttps://www.joltek.com/education/ot-networking-fundamentalsTimestamps00:00:00 Welcome back and the hardware focused modernization theme00:01:40 Benson Hougland background, entrepreneur to controls to Opto 2200:04:10 A garage manufacturing story and the lessons of building real product00:09:00 The gap between cutting edge plants and manual, siloed operations00:11:10 What actually blocks modernization, capital, planning, and alignment00:13:10 Start small, solve a real problem, and build trust in outcomes00:14:40 Proof of concept purgatory and why leadership buy in changes everything00:17:50 Industrial AI needs data, and data integrity becomes the non negotiable00:22:30 Obsolescence, cybersecurity, and simplifying the industrial tech stack00:28:20 Cybersecurity is a process, not a product, and why defaults are deadly00:37:10 Linux at the edge, containers, and why modern controllers are like smartphones00:53:10 ProveIt and the virtual factories approach, real data, real integration paths

Technology modernization in manufacturing is not a list of shiny tools. It is a sequencing problem. In this episode of Manufacturing Hub, Vlad Romanov and Dave Griffith break down why the executive vision for AI often collides with the reality of the plant floor, and what a practical path forward actually looks like when you account for data quality, legacy controls, networking, and the true cost of integration.A core theme in this conversation is imperfect information. Leaders often believe the data already exists because reports exist. But a stack of paper, a few spreadsheets, or a single counter value is not the same as contextualized, trustworthy history that can drive decisions or support advanced analytics. Vlad and Dave walk through why foundational work matters, what teams usually miss during modernization, and how quickly the bill grows when you discover your architecture is outdated, undocumented, or full of dependencies you cannot see until you open panels and start tracing signals.You will also hear a grounded debate on how to think about SCADA, MES, historians, dashboards, and what it would actually mean to “feed data into AI” in a manufacturing context. The takeaway is simple. If you want better outcomes, you need a better understanding of your current state, a clear business case, and a roadmap that prioritizes what matters operationally. Modernization is not one big upgrade. It is a series of decisions that either reduce friction or create it.About the hostsVlad Romanov is an industrial automation and manufacturing expert focused on plant assessments, controls and data architecture, IT and OT integration, and workforce upskilling. Vlad has over 10 years of experience across large manufacturers and complex multi site environments, working from PLC and HMI layers up through SCADA, MES, and ERP integration programs. He is the founder of Joltek, where the mission is to help manufacturers modernize safely, build internal capability, and deliver results that actually survive handoff to operations.Learn more about Joltekhttps://www.joltek.comhttps://www.joltek.com/servicesDave Griffith is an industrial automation practitioner and consultant who works closely with manufacturers to modernize legacy environments, improve reliability, and build practical systems that operators and maintenance teams can support. Dave brings a strong perspective on what is feasible in real plants, where uptime, risk, budget, and organizational readiness drive every decision.Timestamps00:00:00 Welcome and why this month is about technology modernization00:02:10 The real problem with “just add AI” in manufacturing00:04:15 Quick background on Vlad and Dave and the work they do00:05:25 The disconnect between the perfect factory vision and the plant floor00:06:25 Vlad on business cases, integration reality, and infrastructure gaps00:09:05 Dave on imperfect information and why reports are not data00:14:35 What executives actually want from AI and why it is often about people constraints00:20:25 How to get there, hardware first, data normalization, and context00:22:05 Vlad on assessments, legacy hardware, and why upgrades get complicated fast00:39:00 New facility planning mistakes and why early decisions lock you in00:45:10 You have the data, now what, OEE baselines, bottlenecks, and root causes00:58:10 Final takeaways, inventory your architecture and treat data like an assetReferences and links mentionedManufacturing Hub Podcasthttps://www.manufacturinghub.liveProveIt Conferencehttps://www.proveitconference.comAutomate Showhttps://www.automateshow.comIgnition Community Conferencehttps://icc.inductiveautomation.comIf you are watching on YouTube, subscribe so you do not miss the rest of this month’s deep dives on hardware, data teams, and practical applications that actually work on real plant floors.

In this episode of Manufacturing Hub, hosts Vlad Romanov and Dave Griffith welcome back Amos Purdy for a wide ranging conversation that connects plant floor reality with SCADA, MES, and the business decisions that actually fund modernization. Amos shares his path from early software and programming work into industrial automation, including building an industrial automation class and lab, leading MES and SCADA efforts, and working across industries where the pace, constraints, and validation expectations can feel like completely different worlds. If you have ever wondered why a solution that looks obvious on a whiteboard takes months or years to land on a production line, this episode breaks down the human, technical, and financial reasons in plain terms.A big thread throughout the conversation is what it takes to build systems that last. The group digs into hiring and mentoring for Ignition based teams, what backgrounds translate well, and why “hobbyist energy” can be a real superpower in interviews and on the job. The practical takeaway is simple: credentials help you get in the door, but projects help you stand out, especially when you can explain the problem, the architecture, and the tradeoffs you made. The conversation also gets real about legacy plants, where the constraint is often not ambition but risk, ROI, and operational disruption. The group frames modernization as a sequence of targeted moves that improve data availability, reduce cybersecurity exposure, and create a foundation for future applications without betting the entire facility on a massive rip and replace.You will also hear a grounded take on AI in industrial settings. The panel separates what is useful today from what is still hype, and explains why industrial AI needs context, standards, and purpose built training data to be trusted. They connect that to the “data transparency” problem: companies want answers faster, but the hard part is making the data accessible, reliable, and safe in the first place. The episode closes with a discussion on EV and battery manufacturing trends, the reality of global standards and certification, and what the next few years could look like as edge devices, connectivity, and power systems evolve.HostsVlad Romanov is an industrial automation and manufacturing systems expert focused on SCADA, MES, OT data infrastructure, and modernization strategy. He combines electrical engineering depth with an MBA from McGill University to help manufacturers reduce risk, improve reliability, and turn plant data into decision ready information. He leads Joltek, where he delivers assessments, integration roadmaps, and practical upskilling for engineering and operations teams.Dave GriffithManufacturing and automation leader focused on bridging business outcomes with engineering execution, change management, and scalable plant systems.GuestAmos PurdyMBA and electrical engineering background with deep experience across industrial automation, SCADA, MES, and manufacturing intelligence, including leading teams and deployments in both legacy and greenfield environments.Timestamps00:00 Welcome to Manufacturing Hub and why this episode sets up the upcoming modernization theme02:20 Amos Purdy returns and reintroduces his background03:00 From early programming to industrial automation, lab building, and MES leadership09:40 Switching industries and why vertical experience is often overvalued12:40 Hiring and mentoring for Ignition, web skills vs plant floor instincts16:10 AI vs fundamentals, why legacy tech knowledge still matters17:20 Growing teams and how managers should match work to strengths20:10 How candidates stand out, hobby projects and real systems thinking22:50 Technology modernization, data visibility, and cybersecurity as the forcing function31:50 The real bottlenecks, selling ROI, scoping, and avoiding project blowouts37:30 AI readiness in industry, what works today and what is not there yet41:00 EV and battery manufacturing, investment, standards, and what changes on the shop floor50:40 Predictions for the future, edge devices, connectivity, and more data everywhere54:20 Book recommendation and why macro trends matter for engineers56:00 Where to find Amos and what to reach out aboutReferences and links mentionedIgnition by Inductive Automationhttps://inductiveautomation.com/ignitionIgnition SCADA overviewhttps://inductiveautomation.com/scada-softwareInductive University traininghttps://inductiveuniversity.comProveIt Conference 2026 detailshttps://www.proveitconference.comEdison Motorshttps://www.edisonmotors.ca2030: How Today’s Biggest Trends Will Collide and Reshape the Future of Everything by Mauro F. Guillénhttps://us.macmillan.com/books/9781250772213/2030howtodaysbiggesttrendswillcollideandreshapethefutureofeverything/https://www.joltek.com/serviceshttps://www.joltek.com/education/ot-networking-fundamentals