yt:channel:erjTexarRNjdAv_1tbe49A erjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2019-04-29T18:51:02+00:00 yt:video:KzDaWzY-7MU KzDaWzY-7MU UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-09-20T17:00:10+00:00 2026-09-20T17:00:10+00:00 How to Design an Ontology for Manufacturing Operations In this video, I show you a practical approach to designing an ontology for manufacturing operations. Starting with a small set of foundational concepts and relationships that can grow across production, maintenance, quality, inventory, and other operational domains. 𝐖𝐡𝐚𝐭 𝐲𝐨𝐮’𝐥𝐥 𝐥𝐞𝐚𝐫𝐧: 1. The difference between top-down, bottom-up, and hybrid approaches to ontology design 2. Why trying to model your entire manufacturing operation upfront becomes difficult to maintain and scale 3. How to design a small core ontology that remains stable while individual domains extend it 4. The nine foundational concepts I would use for manufacturing operations: Process, Physical Entity, Operational Entity, Agent, Location, Capability, Role, Event, and State 5. How concepts such as ISA-95 can be used to extend the ontology rather than reinventing everything from scratch 6. How the same core model can be extended differently across production, maintenance, quality, inventory, and other domains 7. How to define relationships such as AFFECTS, PARTICIPATES_IN, PLAYS_ROLE, HAS_CAPABILITY, REQUIRES_CAPABILITY, and IS_GUIDED_BY 8. How events, states, and time-bound assertions allow you to represent what happened and how operational conditions changed over time 9. How processes and locations can be represented hierarchically using relationships such as HAS_SUBPROCESS and CONTAINS 10. How these concepts and relationships become the specification for a manufacturing semantic data layer 𝐈𝐧 𝐭𝐡𝐢𝐬 𝐞𝐩𝐢𝐬𝐨𝐝𝐞, 𝐰𝐞 𝐜𝐨𝐯𝐞𝐫: (00:00) Why manufacturing data needs more than consistent names and data models (00:58) How do you actually design a manufacturing ontology? (01:18) Top-down vs. bottom-up ontology design (03:16) Why I recommend a hybrid approach (03:54) Designing the core manufacturing ontology (04:33) The nine foundational concepts (05:21) Process (05:55) Physical Entity (06:15) Operational Entity (06:51) Agent (07:14) Location and ISA-95 (07:45) Capability (08:07) Role (08:50) Event (09:26) State (10:23) Extending the core ontology across manufacturing domains (11:39) Defining the relationships between concepts (12:47) AFFECTS (13:14) PARTICIPATES_IN and PLAYS_ROLE (13:35) HAS_CAPABILITY and REQUIRES_CAPABILITY (13:58) IS_GUIDED_BY and PRODUCES (14:22) OCCURS_AT and IS_LOCATED_AT (14:43) Events, state changes, and time (15:19) Modeling state history with time-bound assertions (15:36) Processes and sub-processes (16:21) Bringing the core relationship model together (17:19) From semantic model to formal ontology 𝐖𝐡𝐞𝐫𝐞 𝐭𝐨 𝐟𝐢𝐧𝐝 𝐊𝐮𝐝𝐳𝐚𝐢 𝐌𝐚𝐧𝐝𝐢𝐭𝐞𝐫𝐞𝐳𝐚: Website: https://www.industry40.tv/ LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ yt:video:D3J--f6RZvQ D3J--f6RZvQ UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-09-13T17:52:47+00:00 2026-09-17T05:27:57+00:00 Understanding Ontology and Knowledge Graphs for Agentic AI in Manufacturing In this video, I explain what a manufacturing ontology actually is, how it differs from a semantic model and a knowledge graph, and why these technologies could become an important foundation for Industrial AI. 𝐖𝐡𝐚𝐭 𝐲𝐨𝐮’𝐥𝐥 𝐥𝐞𝐚𝐫𝐧: 1. Why so much manufacturing knowledge remains implicit rather than formally defined 2. Why operational data alone isn’t enough for AI agents to reason effectively 3. What a semantic model is and why shared vocabulary matters 4. What an ontology is and how it turns that vocabulary into a machine-readable model 5. The difference between an ontology and a knowledge graph 6. How standards such as ISA-95 can provide useful building blocks for manufacturing semantics 7. How ontologies define classes, properties, relationships, and constraints 8. How a knowledge graph gives AI agents a connected view of real assets, processes, materials, events, and other operational entities 9. How these layers together can provide the context an AI agent needs to investigate manufacturing problems 𝐈𝐧 𝐭𝐡𝐢𝐬 𝐞𝐩𝐢𝐬𝐨𝐝𝐞, 𝐰𝐞 𝐜𝐨𝐯𝐞𝐫: (00:00) The invisible knowledge that keeps manufacturing operations running (02:17) Why this becomes a problem for software and AI agents (02:49) Root-cause investigation as an example of the challenge (04:03) What an ontology actually is (05:03) Semantic models, ontologies, and knowledge graphs (05:33) Building a shared manufacturing vocabulary (06:11) Why standards such as ISA-95 matter (07:44) From semantic model to machine-readable ontology (08:34) The architectural blueprint analogy (10:01) From ontology to knowledge graph (11:13) How ontologies and knowledge graphs help AI agents reason (12:32) What’s coming next: designing a manufacturing ontology This is the first step toward a bigger question I’ll explore throughout this series: 𝐇𝐨𝐰 𝐝𝐨 𝐰𝐞 𝐠𝐢𝐯𝐞 𝐀𝐈 𝐚𝐠𝐞𝐧𝐭𝐬 𝐞𝐧𝐨𝐮𝐠𝐡 𝐮𝐧𝐝𝐞𝐫𝐬𝐭𝐚𝐧𝐝𝐢𝐧𝐠 𝐨𝐟 𝐭𝐡𝐞 𝐨𝐩𝐞𝐫𝐚𝐭𝐢𝐨𝐧𝐚𝐥 𝐰𝐨𝐫𝐥𝐝 𝐭𝐨 𝐫𝐞𝐚𝐬𝐨𝐧 𝐫𝐞𝐥𝐢𝐚𝐛𝐥𝐲 𝐚𝐛𝐨𝐮𝐭 𝐦𝐚𝐧𝐮𝐟𝐚𝐜𝐭𝐮𝐫𝐢𝐧𝐠? 𝐖𝐡𝐞𝐫𝐞 𝐭𝐨 𝐟𝐢𝐧𝐝 𝐊𝐮𝐝𝐳𝐚𝐢 𝐌𝐚𝐧𝐝𝐢𝐭𝐞𝐫𝐞𝐳𝐚: Website: https://www.industry40.tv/ LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ yt:video:gIskrw1azBg gIskrw1azBg UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-05-07T10:29:03+00:00 2026-09-08T06:30:12+00:00 Designing Autonomous Multi Agent Systems for Industrial Operations Designing Multi-Agent Systems for Industrial Operations: Kence Anderson, CEO & Founder of AMESA, joins Kudzai Manditereza on the AI in Manufacturing podcast to explore how machine teaching methodology and multi-agent systems are transforming how manufacturers capture expert knowledge and deploy autonomous decision-making on the plant floor. Podcast links: • Spotify: https://open.spotify.com/episode/2x9eK31xvxuDfxbDvuqrNO?si=27b292c224554823 • Apple: https://podcasts.apple.com/de/podcast/industry40-tv/id1541197224?l=en-GB&i=1000766579539 ## Timestamps 00:00 – Introduction 02:55 – The Biggest Gap Between AI Promise and Reality in Manufacturing 05:39 – The Research-to-PR Pipeline: Why R&D's "Development" Phase Is Missing 08:30 – Real-World Example: Teaching AI to Run a Cheetos Extruder 10:47 – Pilot Purgatory and Misallocated Innovation Bandwidth 12:21 – Four Ways to Make Decisions: Calculate, Search, Look Up, or Learn by Practicing 15:28 – What Is Machine Teaching and How Does It Differ from Traditional AI Approaches 19:30 – Glass Manufacturer Case Study 24:42 – Why Multi-Agent Design Beats Monolithic AI in Manufacturing 29:05 – Multi-Agent Design Patterns: Strategy, Perception, Plan & Execute 34:49 – Digital Twins vs. Agents: Separating Feedback from Action 37:05 – AMESA Platform Overview: Agent Orchestration Studio, Agent Cloud, and Runtime 40:02 – Building Multi-Agent Systems Without Code in the Agent Orchestration Studio 42:50 – Simulation-Driven Training: Data Scores, Simulation Scores, and Faithful Digital Environments 45:30 – Deploying Agents to Production: Edge Runtime, PLCs, and Industrial Protocols 47:53 – Scaling AI Agents Across Plants, Recipes, and Formulas Using Operating Regions 50:49 – Real-World Customers: Chemical Process Control, Logistics, and US Military 52:42 – From Data to Autonomy in 12 Weeks: How to Engage with Amesa 54:03 – Prediction: The Bifurcation of Gigafactories vs. Human-Agent Collaboration AI in Manufacturing Podcast with Kudzai Manditereza ***************** Connect With Me: • LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ • Website: https://www.industry40.tv/ ***************** Guest - Kence Anderson ***************** • LinkedIn: https://www.linkedin.com/in/kence/ • Website: https://www.amesa.com/ Kence Anderson is the founder and CEO of AMESA and former Director of Autonomous AI Adoption at Microsoft. He is a pioneer in the field of intelligent autonomous agents, having co-created “Machine Teaching”, a methodology that enables AI agents to develop real-world autonomy through simulation, feedback, and trial-and-error. Over the past seven years, Kence has focused exclusively on designing, building, and deploying intelligent autonomous agents for manufacturing and logistics, leading over 200 real-world deployments for major corporations, including Shell, PepsiCo, and Delta Airlines. He is also the author of Designing Autonomous AI (O’Reilly, 2022) and is now developing a horizontal platform for orchestrating AI agents to make million-dollar decisions in enterprise operations. ***************** PODCAST SPONSORED BY ***************** HiveMQ: https://www.hivemq.com/ ***************** yt:video:KUw0d23U-iA KUw0d23U-iA UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-04-30T10:00:37+00:00 2026-09-07T21:44:00+00:00 I3X Masterclass: How Industrial Information Interoperability eXchange (i3X) Common API Works How Industrial Information Interoperability eXchange (i3X) Common API Works Matthew Parris, Director of Quality Test Systems at GE Appliances and leading contributor to the I3X specification, joins Kudzai Manditereza on the AI in Manufacturing podcast to explore how the Industrial Information Interoperability Exchange (I3X) common API is solving manufacturing's 20-year data access problem and enabling scalable industrial intelligence. Podcast links: • Spotify: https://open.spotify.com/episode/2L1iQjmJYoM31FC2Ffxr8i?si=44960455c872404b • Apple: https://podcasts.apple.com/de/podcast/industry40-tv/id1541197224?l=en-GB&i=1000764645814 ## Timestamps 00:00 – Introduction 00:58 – Matthew Paris' Background and Role at GE Appliances 03:12 – The Origin Story: Why a Common API for Manufacturing? 06:30 – The Data Access Stack: From Transport to Application Layer 08:55 – The Risk Shift: From Picking the Right Tool to Adapting Continuously 10:23 – Software Proliferation and the Need for a Stable Architecture 14:37 – Moving From Visibility and Dashboards to AI-Driven Intelligence 17:59 – Poll vs. Subscription: Why Dashboards Aren't Enough 21:30 – Building the Foundation for AI Agents in Manufacturing 24:35 – What Is I3X? A Technical Overview of the Specification 30:00 – The I3X Explorer: Netscape for Manufacturing Data 34:49 – I3X vs. OPC UA: What's Different and Why It Matters 41:12 – I3X and the Unified Namespace (UNS): How They Work Together 50:23 – OPC UA Part 5 Information Models and Type Definitions 58:22 – Adoption Strategy: Solving the Chicken and Egg Problem 1:02:15 – How to Get Started with I3X and Final Thoughts AI in Manufacturing Podcast with Kudzai Manditereza ***************** Connect With Me: • LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ • Website: https://www.industry40.tv/ ***************** Guest - Matthew Parris ***************** • LinkedIn: https://www.linkedin.com/in/matthewparris/ • i3x Website: https://www.i3x.dev/ • i3x Github: https://github.com/cesmii/i3X Resources https://www.linkedin.com/pulse/uns-further-explained-matthew-parris-8pjme/ https://www.linkedin.com/pulse/uns-glance-matthew-parris-zc8ne/ ***************** PODCAST SPONSORED BY ***************** HiveMQ: https://www.hivemq.com/ ***************** yt:video:meLba-JdDMI meLba-JdDMI UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-04-22T10:01:22+00:00 2026-09-07T11:09:27+00:00 Optimizing AI Inferencing for Agentic Operations in Manufacturing How to Optimize AI Inference for Agentic Operations in Manufacturing: Kelvin Cooper, Co-Founder and CEO of Neurometric.ai, joins Kudzai Manditereza on the AI in Manufacturing podcast to reveal why routing all AI tasks through a single frontier model becomes a liability at scale—and how inference orchestration with small language models (SLMs) can deliver 10x cost and latency improvements for manufacturers. Podcast links: • Spotify: • Apple: ## Timestamps 00:00 – Introduction 02:29 – Why AI Will Touch Every Workflow and Every Business 03:32 – Why AI Pilots Fail to Reach Production 05:33 – Overcoming the Committee Mindset: Two-Way Doors and Shipping Fast 07:52 – How to Identify Your First AI Use Case in Manufacturing 10:04 – Agentic AI in Manufacturing: Why One Frontier Model Isn't Enough 12:31 – AT&T's Case Study: Cutting AI Costs by 90% with Orchestration 14:13 – Intelligence Gets Headlines, But Reliability Determines Scale 16:27 – Fine-Tuned SLMs vs. General Purpose LLMs: What Makes AI Reliable 18:13 – AI Maturity Framework: Stages of Deployment for Manufacturers 20:32 – Catastrophic Forgetting: The Hidden Challenge in Operationalizing AI 22:46 – Capital Flooding Into Manufacturing AI: Billion-Dollar Rollup Funds 24:16 – The Future Is a Coordinated Team of Models, Not One Model 27:25 – Avoiding Vendor Lock-In: Building AI Systems With Abstraction 28:08 – Neurometric Platform Breakdown: SLM Marketplace and Model Analysis 31:20 – AI Managing AI: Self-Optimizing Infrastructure and Dynamic Orchestration 32:56 – Predictions: The Factory of the Future and Jevons Paradox for Labor AI in Manufacturing Podcast with Kudzai Manditereza ***************** Connect With Me: • LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ • Website: https://www.industry40.tv/ ***************** Guest - Kelvin Cooper ***************** • LinkedIn: https://www.linkedin.com/in/coopernyc/ • Website: https://www.neurometric.ai/ • Neurometric Substack: https://neurometric.substack.com/ ***************** PODCAST SPONSORED BY ***************** HiveMQ: https://www.hivemq.com/ ***************** yt:video:8goUIlTKxno 8goUIlTKxno UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-04-01T10:12:24+00:00 2026-09-07T13:37:22+00:00 How to Build AI Solutions That Actually Work on the Factory Floor Renan Devillieres, Founder & CEO of OSS Ventures, joins Kudzai Manditereza on the AI in Manufacturing podcast to reveal what it actually takes to build and scale AI from pilot to hundreds of factories, why 95% of factories are falling behind tech-enabled competitors, and how his venture studio has deployed AI across 3,800+ factories. --- Podcast links: • Spotify: https://open.spotify.com/episode/22ySvYFExkv3vi8iW99PJJ?si=77d995e0e68340d7 • Apple: https://podcasts.apple.com/de/podcast/industry40-tv/id1541197224?l=en-GB&i=1000758601509 --- **TIMESTAMPS:** 00:00 – Introduction r 03:58 – What Is OSS Ventures? Building & Spinning Out 22 Manufacturing Startups 04:43 – Why Only 5% of Factories Operate Like Tech Companies 06:40 – The Skills Gap: Why Digitization Leaders Need to Understand Code 08:32 – Thinking in Systems, Not Parts: Lessons from Tesla, Xiaomi & BYD 10:13 – Solving Manufacturing's Talent Attraction & Image Problem 13:29 – The Historical Parallel: AI as the New Industrial Revolution 16:35 – Industrializing Discovery: Why 85% of AI Projects Fail 20:43 – Key Learnings from Deploying AI in 100+ Factories 25:52 – The OSS Ventures Validation Process: The 10x Rule & Getting Factory Directors to Pay 26:42 – Top AI Use Cases: How to Spot Genuine Pain Points on the Shop Floor 30:45 – Designing Usable AI: Why Managers of AI Agents Need Better UX 34:09 – Embedding AI vs. Building a New Layer: Why Copilots Aren't Enough 35:57 – Scaling AI from Zero to 600+ Factories: Product Love vs. Military-Grade Deployment 39:07 – Reusable Tech Bricks: The Role of Shared Infrastructure in Scaling 40:41 – How to Work with OSS Ventures: For Founders & Factory Directors 42:41 – Prediction: Manufacturing Pay Rises 25% & MIT Grads Flock to Factories --- AI in Manufacturing Podcast with Kudzai Manditereza ***************** Connect With Me: • LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ • Website: https://www.industry40.tv/ ***************** Guest - Renan Devillieres ***************** • LinkedIn: https://www.linkedin.com/in/renan-devillieres/ • Website: https://www.oss.ventures/ ***************** PODCAST SPONSORED BY ***************** HiveMQ: https://www.hivemq.com/ ***************** yt:video:ugFJMBeu2uE ugFJMBeu2uE UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-03-25T11:01:32+00:00 2026-09-07T18:38:47+00:00 How to Build and Scale Agentic AI Workflows in Manufacturing with Causal AI Bernard Kratzwald, Co-Founder and CTO at Ethon AI, joins Kudzai Manditereza on the AI in Manufacturing podcast to explore how process knowledge graphs, causal AI, and purpose-built model layers enable manufacturers to move from correlation-based analytics to truly autonomous, scalable agentic workflows on the shop floor. --- Podcast links: • Spotify: https://open.spotify.com/episode/2hLrGJyXmcnLzQNmIIkAL1?si=0e2734aa10454acc • Apple: https://podcasts.apple.com/de/podcast/industry40-tv/id1541197224?l=en-GB&i=1000757225194 --- ## Timestamps 00:00 – Introduction 03:51 – Core Business Objectives Driving Manufacturers Toward Agentic AI 06:25 – Why the Current Industrial Data Stack Falls Short for AI 08:38 – What Contextualization Really Means on the Factory Floor 11:56 – Process Knowledge Graphs vs. Traditional Data Models 14:52 – Monolithic vs. Federated Knowledge Graph Approaches 17:58 – Causal AI vs. Correlation-Based Analytics in Manufacturing 20:32 – Architecture Layers for Agentic Workflows at Scale 24:50 – Why Purpose-Built Industrial Models Beat General-Purpose LLMs 27:43 – Change Management, Trust, and Explainability on the Shop Floor 29:44 – Ethon AI Platform Breakdown: From Monitoring to Autonomous Control 32:26 – AI-Native Historian and MES Capabilities 34:32 – Capturing Expert Knowledge into the Process Knowledge Graph 41:00 – Integration with Existing IT/OT Systems (MQTT, Kafka, SAP, SQL) 43:12 – Governance, Security, and Compliance for Enterprise AI 47:48 – Deployment Models: SaaS, Private Cloud, and Edge 48:03 – Engagement Timeline: From Kickoff to First Value in Under 3 Months 50:48 – Case Studies: Siemens and Lindt & Sprüngli 53:40 – Predictions: Autonomous Control, Cross-Factory Intelligence & Workforce Transformation --- AI in Manufacturing Podcast with Kudzai Manditereza ***************** Connect With Me: • LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ • Website: https://www.industry40.tv/ ***************** --- Guest - Bernard Kratzwald ***************** • LinkedIn: https://www.linkedin.com/in/bernhard-kratzwald/ • Website: https://www.ethon.ai/customers • World Economic Forum report: https://reports.weforum.org/docs/WEF_Proof_over_Promise_Insights_on_Real_World_AI_Adoption_from_2025_MINDS_Organizations_2026.pdf ***************** --- PODCAST SPONSORED BY ***************** HiveMQ: https://www.hivemq.com/ ***************** yt:video:GZPWih8uE-M GZPWih8uE-M UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-03-17T10:33:17+00:00 2026-09-08T06:18:02+00:00 Why Unified Namespace is THE Essential Foundation for Industrial AI and Agentic Operations Walker Reynolds, President and Solutions Architect at 4.0 Solutions and founder of the ProveIt Conference, joins Kudzai Manditereza on the AI in Manufacturing podcast to share his top takeaways from Prove It 2025, reveal why knowledge graphs are the next critical skill for manufacturers, break down what separates UNS success from failure, and lay out his ideal full-stack industrial data architecture for the AI era. Podcast links: • Spotify: https://open.spotify.com/episode/6Trp49ZNW5kEpAxmH5V8jG?si=e10c67d71aad4a54 • Apple: https://podcasts.apple.com/de/podcast/industry40-tv/id1541197224?l=en-GB&i=1000755723999 00:00 – Introduction 07:59 – State of the Industry: Three Key Observations from ProveIt 2026 10:45 – Why Knowledge Graphs Are a Game Changer for Manufacturing AI 15:31 – Knowledge Graphs Will Be Native to Every Platform 17:20 – Federated Knowledge Graphs & Bottom-Up Ontology Building 18:09 – Standardized vs. Ad Hoc Relationships in Knowledge Graphs 20:03 – Top 5 Practical AI Solutions from Prove It 2025 29:08 – Why Not Everyone Can Make the Leap to Agentic AI 31:15 – AI Agents Need a Backbone: Why Context Across Systems Matters 33:43 – The Pattern: AI Is Best at Building & Why Agents Aren't Autonomous Yet 42:11 – Unified Namespace: What Separates Success from Failure 48:23 – UNS Is Current State — Stop Trying to Make It Something It's Not 51:16 – UNS as the Natural Backbone for Agentic AI 54:40 – Walker's current Full-Stack Industrial Data Architecture and tools 59:25 – Where Does AVEVA PI Fit in a UNS Architecture? 1:02:11 – Prediction AI in Manufacturing Podcast with Kudzai Manditereza ***************** Connect With Me: • LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ • Website: https://www.industry40.tv/ ***************** Guest - Walker Reynolds ***************** • LinkedIn: https://www.linkedin.com/in/walkerdreynolds/ • Website: https://www.iiot.university/ • Unified namespace: https://virtualfactory.online/ ***************** PODCAST SPONSORED BY ***************** HiveMQ: https://www.hivemq.com/ ***************** yt:video:tfSbkcaoyDY tfSbkcaoyDY UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-03-11T11:01:38+00:00 2026-09-08T06:13:05+00:00 Unlocking Productivity in Manufacturing With Casual Models and Agentic AI Michael Carroll, Global Executive in Industrial Innovation & AI, Strategic Advisor and Fellow, COO Council at LNS Research, and Chief Strategy Officer at TrekAI, joins Kudzai Manditereza on the AI in Manufacturing podcast to unpack why a casual model infrastructure and agentic AI represents a fundamental structural shift and how manufacturers can close the persistent productivity gap by rethinking decisions, permissions, and organizational architecture. **Podcast links:** • Spotify: https://open.spotify.com/episode/3kFxvNrF4h90tCohCa7MLJ?si=6cf06b7cfc874df8 • Apple: https://podcasts.apple.com/de/podcast/industry40-tv/id1541197224?l=en-GB&i=1000754620914 ## Timestamps 00:00 – Introduction and Guest Welcome 04:04 – Why Manufacturing Productivity Has Been Flat Since 2008 06:43 – Georgia Pacific's Digital Transformation Journey 09:39 – Why the Knowledge Loss Hypothesis Doesn't Fully Explain the Productivity Decline 11:16 – The COVID Productivity Spike and the General Mills Lesson on Focus 13:41 – The Cognitive Tipping Point: Why More Insights Don't Equal Better Performance 15:41 – What Is Agentic AI and How Does It Differ From Traditional Software? 18:59 – Enterprise Agency: Decisions as the Atomic Unit of Work 22:46 – Causal Reasoning vs. Explainable AI: Why the Difference Matters 27:13 – What Changes for the Shop Floor Operator With Agentic AI 28:32 – From Fixed Architecture to Adaptive Architecture 33:47 – Rethinking the Operating Model: Collapsing the Coordination Middle Layer 37:33 – Decision Speed and Performance: The J. Robert Baum Study 39:06 – Assumption Observability: Treating Beliefs as First-Class Entities 42:26 – Bridging the OT-IT Divide: Enablement Over Permission 46:28 – Why Causal Models Beat Data Models for Manufacturing 49:44 – Knowledge Graphs, Causal Graphs, and Chains of Reasoning 52:04 – Where to Begin: Mapping Inferencing Load and Permission Load 56:37 – Join the Conversation: We're Not Trying to Be Right, We're Trying to Get This Right ## AI in Manufacturing Podcast with Kudzai Manditereza ***************** Connect With Me: • LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ • Website: https://www.industry40.tv/ ***************** ## Guest - Michael Carroll ***************** • LinkedIn: https://www.linkedin.com/in/michael-carroll-0106367/ • Website: https://www.lnsresearch.com/ ***************** ## PODCAST SPONSORED BY ***************** HiveMQ: https://www.hivemq.com/ ***************** yt:video:TU_iDQ9BnZc TU_iDQ9BnZc UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-03-05T11:00:45+00:00 2026-06-24T18:03:11+00:00 Context Engineering Techniques for Building Reliable Industrial AI Agents Zach Etier, VP of Architecture at Flow Software, joins Kudzai Manditereza on the AI in Manufacturing podcast to unpack context engineering techniques for building reliable industrial AI agents and the role of knowledge graphs in scaling operational intelligence. Podcast links: • Spotify: https://open.spotify.com/episode/2kBSPIS2YIfKIvPc1IEUVE?si=8ceadae789f44403 • Apple: https://podcasts.apple.com/de/podcast/industry40-tv/id1541197224?l=en-GB&i=1000753193433 ## Timestamps 00:00 – Introduction 05:21 – Defining AI Agents for the Industrial Context 09:30 – Industrial Agent Example: Automated Shift Handover Reports 11:07 – From Prompt Engineering to Context Engineering 14:45 – Why More Data Can Actually Hurt Agent Performance 18:59 – Types of Context an Industrial Agent Needs 24:19 – Key Terms Explained: MCP, Skills, Sub-Agents & Context Rot 29:51 – Three Context Engineering Techniques 36:44 – Skills vs. MCP vs. Sub-Agents: When to Use Each 42:50 – How Skills Capture SOPs and Enable Modular Agent Design 49:31 – Building Reliable Industrial Agents: A Repeatable Workflow 54:21 – Data Architecture for Enterprise-Scale AI: Knowledge Graphs & Ontologies 1:00:35 – Federated Knowledge Graphs vs. Top-Down Enterprise Modeling 1:05:03 – Flow Software's Platform: Timebase, Atlas & the AI Gateway 1:10:33 – Prediction: The Future of Agents on the Factory Floor AI in Manufacturing Podcast with Kudzai Manditereza ***************** Connect With Me: • LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ • Website: https://www.industry40.tv/ ***************** Guest - Zach Etier ***************** • LinkedIn: https://www.linkedin.com/in/zach-etier-a90b4840/ • Workshop: https://github.com/Zach-etier/ProveIT_2025_Flow_Workshop • Flow Software Website: https://www.flow-software.com/ ***************** PODCAST SPONSORED BY ***************** HiveMQ: https://www.hivemq.com/ ***************** yt:video:BB7CLMHkN3w BB7CLMHkN3w UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-02-26T10:00:15+00:00 2026-09-07T07:14:04+00:00 Multi AI Agent Based Quality Control in Manufacturing: Reducing Waste and Improving Efficiency Willem Klein, CEO and Co-Founder of Zetamotion, joins Kudzai Manditereza on the AI in Manufacturing podcast to explore how AI-powered visual inspection is being democratized for manufacturers — reducing weeks of manual data labeling to under an hour and making production-grade quality control accessible without a data science team. Podcast links: • Spotify: https://open.spotify.com/episode/0HAEDHAUP173f9q08BcWDT?si=1511f333b6a54631 • Apple: https://podcasts.apple.com/de/podcast/industry40-tv/id1541197224?l=en-GB&i=1000751728999 ## Timestamps 00:00 – Introduction 03:28 – Where Zetamotion Fits in the AI Manufacturing Landscape 05:01 – The History of AI Adoption and the GPT Moment for Industry 08:30 – Why Over 90% of Industrial AI Pilots Fail 11:06 – Shadow AI: Unsanctioned Projects Driving Real Innovation 14:48 – Balancing AI Governance with Flexibility on the Factory Floor 18:08 – Why System-Level Thinking Beats a Better AI Model 21:44 – Introducing Zelia: The End-to-End AI Inspection Assistant 26:36 – How Zelia and Spectron Work Together 28:27 – The Full Vision: Fully Autonomous Inspection Setup by End of Year 30:03 – The Role of Human Feedback in AI-Powered Quality Control 33:46 – Time Savings: From 100,000 Labeled Images to Five Samples 35:51 – Edge vs. Cloud Deployment for Manufacturing AI 38:10 – Scaling Challenges: Why No Two Production Lines Are the Same 42:44 – The Bigger Vision: Physical AI and Beyond Defect Inspection 45:48 – Prediction: Major Automation Breakthroughs Are Imminent AI in Manufacturing Podcast with Kudzai Manditereza ***************** Connect With Me: • LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ • Website: https://www.industry40.tv/ ***************** Guest - Willem Klein ***************** • LinkedIn:https://www.linkedin.com/in/wilhelm-e-j-klein/ • Website: https://zetamotion.com/ • Website: https://zetamotion.com/zetamotion-end-to-end-learning-inspection-assistant/ ***************** PODCAST SPONSORED BY ***************** HiveMQ: https://www.hivemq.com/ ***************** yt:video:ns8L5j6AQxA ns8L5j6AQxA UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-02-19T11:01:19+00:00 2026-09-08T23:15:27+00:00 A Practical Guide to Implementing Industrial AI Agents in Factories Practical Guidance for Implementing Industrial AI Agents in Manufacturing: James Zhang, Co-Founder and Chief Product Officer of OpsMate AI, joins Kudzai Manditereza on the AI in Manufacturing podcast to share how agentic AI creates a new decision intelligence layer that augments skilled workers and solves the manufacturing productivity plateau. --- Podcast links: • Spotify: https://open.spotify.com/episode/3gxwg3GPtOs3krPnSBEgyB?si=faa02fff9f2549d0 • Apple: https://podcasts.apple.com/de/podcast/industry40-tv/id1541197224?l=en-GB&i=1000750453366 --- Timestamps 00:00 – Introduction 03:53 – Why Manufacturers Should Care About Agentic AI 04:38 – The Manufacturing Productivity Plateau and the Skilled Labor Crisis 08:50 – How Agentic AI Creates Digital Workers to Augment Factory Teams 11:05 – The Decision Intelligence Layer vs. Adding Copilots to Existing Systems 15:55 – Why AI Agents Won't Replace Legacy Systems Like MES and ERP 17:46 – Identifying High-Value Use Cases for Agentic AI in Factories 24:07 – Do You Need Perfect Data Infrastructure for AI Agents? 30:00 – Context Graphs vs. Knowledge Graphs: The Foundation for Industrial AI Agents 37:21 – Deep Dive Into the OpsMate AI Platform Architecture 45:35 – Where OpMate Sits in a Typical Manufacturing Data Stack 47:55 – Real Customer Examples: Automotive, Discrete Manufacturing, and ETO 52:40 – The Future of Manufacturing Work 56:12 – Advice for Manufacturing Leaders: Top-Down Governance Meets Bottom-Up Innovation --- AI in Manufacturing Podcast with Kudzai Manditereza ***************** Connect With Me: • LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ • Website: https://www.industry40.tv/ ***************** Guest - James Zheng ***************** • LinkedIn: https://www.linkedin.com/in/jameszhangboston/ • Website: https://www.opsmateai.com/ ***************** PODCAST SPONSORED BY ***************** HiveMQ: https://www.hivemq.com/ ***************** yt:video:F0oaVkVj2EQ F0oaVkVj2EQ UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-02-03T11:00:20+00:00 2026-09-08T03:48:59+00:00 How Manufacturers Scale from Fragmented Data to AI-Native Intelligence: Data Layer, UNS, MCP, I3X Building a Foundation for AI-Native Industrial Intelligence: Craig Scott, CEO and Founder of Fuuz, joins Kudzai Manditereza on the AI in Manufacturing podcast to reveal why most industrial AI initiatives fail and how a model-driven approach creates the data foundation manufacturers need to scale from pilot to production. Podcast links: • Spotify: https://open.spotify.com/episode/3p67oH9s45xPfUm5NuAEws?si=17e3b0dd155949a7 • Apple: https://podcasts.apple.com/de/podcast/industry40-tv/id1541197224?l=en-GB&i=1000747829480 TIMESTAMPS 00:00 – Introduction 01:26 – Craig's Journey From Shop Floor to Industrial Intelligence Platform Founder 06:56 – What's Fundamentally Broken in Manufacturing Data Architecture 08:09 – Why Real-Time Shop Floor Data Never Reaches Enterprise Systems 16:20 – Enabling AI With i3x, MCP, and GraphQL Standards 18:52 – Model-Driven Approach vs Point Solutions for Data Integration 23:14 – Balancing Data Modeling Rigor With Speed to Value 26:27 – Why AI Governance Requires Deterministic Data Models 28:41 – ISA-95 Standards vs Custom Data Models in Practice 33:14 – Red and Blue Namespace: IT Governance Meets OT Flexibility 37:28 – What Is Fuuz? MES, WMS, and Operational Intelligence Platform 41:53 – Monolithic Systems vs Best-in-Class Tools Debate 46:46 – Fuuz Platform Architecture: MongoDB, Kubernetes, and React Stack 51:07 – Real-World Deployments: Automotive OEM, Steel Mills, and CPG 53:52 – How to Prepare Your Manufacturing Data for AI-Native Operations AI in Manufacturing Podcast with Kudzai Manditereza ***************** Connect With Me: • LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ • Website: https://www.industry40.tv/ ***************** Guest - Craig Scott ***************** • LinkedIn: https://www.linkedin.com/in/craigascott1/ • Website links: https://www.fuuz.com/ https://support.fuuz.com https://academy.fuuz.com/ ***************** PODCAST SPONSORED BY ***************** HiveMQ: https://www.hivemq.com/ ***************** yt:video:ImLPaZvG_2M ImLPaZvG_2M UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2026-01-22T11:00:35+00:00 2026-09-07T05:58:16+00:00 Driving Operational Excellence in Manufacturing with Practical AI Mickey Shaposhnik, CEO of Next Plus, reveals why traditional MES is dead and how AI-powered manufacturing execution is helping factories capture tribal knowledge, eliminate paper-based processes, and deploy electronic batch records in weeks instead of years. Giveaway to Industry40tv listeners 1-Year Free Access to Next Plus: https://nextplus.io/industry40-tv-giveaway/ 𝗣𝗼𝗱𝗰𝗮𝘀𝘁 𝗹𝗶𝗻𝗸𝘀: • Spotify: https://open.spotify.com/episode/3f1452hqyGXoFaCvca4hPu?si=960d2dbe7dc44ffa • Apple: https://podcasts.apple.com/de/podcast/industry40-tv/id1541197224?l=en-GB&i=1000746191930 𝗧𝗶𝗺𝗲𝘀𝘁𝗮𝗺𝗽𝘀: 0:00 - Introduction: AI-Powered Manufacturing Execution 2:51 - Why Paper-Based Manufacturing Is Failing in 2025 5:07 - Data Collection Challenges on the Shop Floor 6:02 - The "Garbage In, Garbage Out" Problem with Digital Forms 8:23 - AI Voice-to-Form: Eliminating Data Entry Friction 9:37 - Bridging the Knowledge Gap Between Business and Shop Floor 13:19 - Why "Training" Is Dead in Manufacturing 15:03 - Building Your Factory's Internal YouTube & Wikipedia 16:28 - The Four Pillars of AI-Driven Manufacturing Execution 17:09 - Digital Work Instructions, Data Collection, Analytics & SOP Creation 20:27 - Why the MES Monolith Is Broken 21:46 - Low Volume, High Mix: Why Agility Beats Rigidity 23:56 - How AI Learns and Generates Insights on the Shop Floor 24:30 - Fault Reports & AI Agents: 20-50% Reduction in Expert Calls 28:54 - Humanoid Robots Still Need Tribal Knowledge 29:49 - IT/OT Integration: MQTT, OPC UA & Open Architecture 30:34 - Three Cybersecurity Architectures for MES Deployment 33:24 - Scaling Globally: How AI Eliminates the $100K Translation Problem 38:02 - Real-World Use Cases: Aerospace to Pharma 38:28 - Electronic Batch Records (EBR): From 2 Years to 6 Weeks 40:49 - 2015 Mindset vs. 2025 Technology Capabilities 41:22 - The COVID Vaccine Lesson: Speed of AI Transformation 42:24 - Skip Pilots, Start Small Projects: Advice for Manufacturing Leaders 43:24 - Where to Find Miki Shaposhnik AI in Manufacturing Podcast with Kudzai Manditereza ********************* Connect With Me: • LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ • Website: https://www.industry40.tv/ ******************** Guest - Mickey Shaposhnik ******************* • Website: https://nextplus.io/ • Mickey on LinkedIn - https://www.linkedin.com/in/sakranolog/ ******************* PODCAST SPONSORED BY HiveMQ: https://www.hivemq.com/ yt:video:ITPQ16-9W7Y ITPQ16-9W7Y UCerjTexarRNjdAv_1tbe49A Industry40tv https://www.youtube.com/channel/UCerjTexarRNjdAv_1tbe49A 2025-12-24T07:00:31+00:00 2026-09-07T15:01:14+00:00 You don't have a data problem, you have a context problem This video explores why presented data frequently remains unused, primarily because individuals struggle with **understanding data** in a broader context of learning and improvement. The discussion emphasizes that a lack of crucial **context clues** impedes effective **data interpretation**, making it challenging to **explain data** meaningfully. Ultimately, improving **data literacy** is essential for deriving true value from information. AI in Manufacturing Podcast with Kudzai Manditereza *************** Watch/Listen to full episode here: YouTube - https://youtu.be/V7RVDiSt0xQ Spotify - https://open.spotify.com/episode/6PwbGkxCrUKeaKnjUgQcKN?si=1eda9ba1e8b64b5f Apple - https://podcasts.apple.com/de/podcast/industry40-tv/id1541197224?l=en-GB&i=1000727162892 *************** Connect With Me: *************** • LinkedIn: https://www.linkedin.com/in/kudzaimanditereza/ • Website: https://www.industry40.tv/ *************** Guest - Bob van de Kuilen ************ 💼 Bob LinkedIn - https://www.linkedin.com/in/bob-van-de-kuilen-a531403/ 🌐 Thred Website - https://www.thred.cloud/ ************