The Agentic Mesh Podcast

#AICoding #AIAgents #SoftwareEngineering #DeveloperProductivityAgent memory is becoming a new foundation of enterprise agent architecture. But it seems like we are still treating memory as a bigger context window, a retrieval layer, or a better prompt. Instead, making your agents work well is turning out to be based upon how you manage its memory.  In fact, it looks like memory is turning into almost an operating system for agent work: what the agent knows, what it is allowed to see, what it carries through execution, what it shares, and what becomes durable over time.Out Now!  O'Reilly Agentic Mesh written by Eric Broda and Davis Broda👉https://www.oreilly.com/library/view/agentic-mesh/9798341621633/👉Stay Connected with Eric and John:https://www.linkedin.com/in/ericbroda/https://www.linkedin.com/in/jymiller/👉For full-length articles by Eric Broda and John Miller:https://agenticmesh.substack.com/?utm_campaign=profile_chips

Coding agents are already highly effective in the environment they were designed around: local context, local memory, local files, immediate feedback, and bounded technical surfaces. The tools are not a disappointment. They are a genuine shift in how contained technical work gets done.The gap between that and enterprise-scale production delivery is real and will not be closed by better model output alone. It requires better decomposition practices, stronger context engineering, explicit acceptance criteria, hybrid review models, clearer boundary definitions for business users, and enterprise harnesses built for governed participation rather than local convenience. Organizations that do that work will get compounding value. Organizations that skip it will get excellent demos, uneven systems, and hard lessons at the production boundary.☑️Subscribe to our Channel for the most up to date information on Agentic Mesh!Out Now!  O'Reilly Agentic Mesh written by Eric Broda and Davis Broda👉https://www.oreilly.com/library/view/agentic-mesh/9798341621633/👉Stay Connected with Eric and John:https://www.linkedin.com/in/ericbroda/https://www.linkedin.com/in/jymiller/👉For full-length articles by Eric Broda and John Miller:https://agenticmesh.substack.com/?utm_campaign=profile_chips#AICoding #AIAgents #SoftwareEngineering #DeveloperProductivity

APA is the form factor for the enterprise agent ecosystem.Enterprises won't deploy one agent. They'll deploy millions — across functions, vendors, and models. The question isn't whether agents can reason or use tools. The question is what architecture makes them manageable, deployable, and governable at that scale. Agentic Process Automation is that architecture.Out Now!  O'Reilly Agentic Mesh written by Eric Broda and Davis Broda👉https://www.oreilly.com/library/view/agentic-mesh/9798341621633/👉Stay Connected with Eric and John:https://www.linkedin.com/in/ericbroda/https://www.linkedin.com/in/jymiller/👉For full-length articles by Eric Broda and John Miller:https://agenticmesh.substack.com/?utm_campaign=profile_chipsagentic ai, agentic process automation, ai agents, enterprise ai, ai architecture, multi agent systems, agent orchestration, ai workflow automation, business process automation, intelligent automation, ai in enterprise, llm agents, ai agents architecture, enterprise agents, agent ecosystems, ai systems design, workflow automation ai, ai reasoning agents, process automation ai, ai infrastructure, scalable ai systems, agent collaboration, enterprise automation

An agent harness is the runtime system that turns a model into an operating agent, but the harness required for a coding or personal agent is fundamentally different from the one required for an enterprise agent. Coding harnesses are built for workspace execution; enterprise harnesses are built for governed participation in long-running business processes under identity, policy, and trust.Out Now!  O'Reilly Agentic Mesh written by Eric Broda and Davis Broda👉https://www.oreilly.com/library/view/agentic-mesh/9798341621633/👉Stay Connected with Eric and John:https://www.linkedin.com/in/ericbroda/https://www.linkedin.com/in/jymiller/👉For full-length articles by Eric Broda and John Miller:https://agenticmesh.substack.com/?utm_campaign=profile_chips

If Knowledge Management is the library, the Agentic Knowledge Fabric is the precision-curated briefing document prepared specifically for the agent’s current task. The Agentic Knowledge Fabric offers an architecture for token-aware knowledge serving.Out Now! O'Reilly Agentic Mesh written by Eric Broda and Davis Broda👉https://www.oreilly.com/library/view/agentic-mesh/9798341621633/👉Stay Connected with Eric and John:https://www.linkedin.com/in/ericbroda/https://www.linkedin.com/in/jymiller/👉For full-length articles by Eric Broda and John Miller:https://agenticmesh.substack.com/?utm_campaign=profile_chips

Agentic Process Automation puts agents inside enterprise workflows at scale — thousands of them, making decisions, routing work, invoking tools. Evaluating whether each one is safe to operate requires matching performance criteria to the governance risk each agent actually carries.That matching is the core problem. An agent can complete tasks quickly while violating policy. It can remain compliant while consuming too much compute to be viable. It can appear accurate while using tools incorrectly, escalating poorly, or behaving inconsistently across similar cases. A few summary metrics or anecdotal feedback are not enough. And even when the right things are measured, the meaning of a weakness depends on context. A reliability issue that is tolerable in a low-risk drafting workflow is disqualifying in a multi-step process where the agent has broad discretion. Without a way to classify the operating context and then evaluate the agent within that context, organizations drift toward either over-control or under-control.☑️Subscribe to our Channel for the most up to date information on Agentic Mesh!You can also listen to our Podcast on Spotify: https://open.spotify.com/show/6C6U2fmVdxNMdo1bZpVASy?si=3cyoAireSZiBaKJ-1FtJlAApple Podcasts: https://podcasts.apple.com/us/podcast/the-agentic-mesh-podcast/id1874331081Out Now!  O'Reilly Agentic Mesh written by Eric Broda and Davis Broda👉https://www.oreilly.com/library/view/agentic-mesh/9798341621633/👉Stay Connected with Eric and John:https://www.linkedin.com/in/ericbroda/https://www.linkedin.com/in/jymiller/👉For full-length articles by Eric Broda and John Miller:https://agenticmesh.substack.com/?utm_campaign=profile_chips

Agentic Process Automation has agents participating directly in business processes, making step-level decisions, interpreting mixed inputs, coordinating across systems, and operating within policy and control boundaries. That shift matters because many enterprise processes now depend on judgment over documents, messages, exceptions, thresholds, and domain rules that do not fit cleanly into deterministic flow logic.When an agent executes one of those enterprise steps, output quality depends on whether it receives the right context for that specific task: the right subset of enterprise knowledge, containing the correct definitions, policy constraints, exceptions, and decision thresholds. In most enterprises, that context is fragmented across policy manuals, standard operating procedures, source systems, regulatory texts, tickets, emails, and human judgment. As a result, the knowledge needed for a single decision step is rarely assembled in a form that is complete, scoped, and usable at execution time.The Agentic Knowledge Fabric (AKF) is the knowledge foundation for Agentic Process Automation (APA). It addresses that operating gap by converting fragmented enterprise knowledge into bounded context artifacts that can be retrieved and assembled under explicit context limits. AKF is built on an engineering premise: context should be treated as a product with predictable size, stable identifiers, provenance, and deterministic assembly. That premise drives both the logical architecture—how meaning is represented, indexed, linked, and selected—and the operational pipeline that builds and maintains those representations.

What if agents could do more than assist developers — what if they became trusted participants in enterprise business processes?In this episode, we explore Agentic Process Automation (APA) and why it’s the next step beyond personal AI agents. While coding agents operate under individual credentials and user interfaces, enterprise agents must run headless, integrate across SaaS platforms, and meet strict governance, identity, and trust requirements.We discuss:Personal vs. enterprise agentsWhy brittle integrations and hard-coded logic fall shortHow agents can fill the “white space” between systems like Salesforce, Workday, and ServiceNowThe importance of knowledge engineering and shared business contextWhy “Know Your Agent” (KYA) and agent governance are essential at scaleAgents aren’t just automation tools — they can reason, handle ambiguity, and operate 24/7 within structured business processes. But doing that safely requires new architectural thinking.If you're building enterprise AI systems, modernizing integrations, or exploring multi-agent ecosystems, this episode is for you.

We’re entering an era where enterprises won’t run a handful of agents, but thousands or millions of them. In this episode, we introduce Know Your Agent (KYA) and argue that managing agents must look more like managing employees than managing software. We break down why identity, access control, auditability, and trust are now urgent problems, how lessons from KYC, HR, and data governance apply to agents, and why scaling agents without these foundations is a serious risk. If agents are moving at machine speed, trust can’t be an afterthought.

Most enterprise automation fails in the same place: the brittle gap between SaaS systems.In this episode of The Agentic Mesh Podcast, John Miller and Eric Broda argue that traditional RPA and workflows can’t handle ambiguity, edge cases, or real-world complexity. Instead, they explore agentic process automation, where AI agents powered by large language models fill the “white space” between platforms like Salesforce, ServiceNow, and Workday.They break down why knowledge engineering, context, identity, and emerging standards like MCP and A2A are now mandatory for building enterprise-grade agents — not optional.If you’re still trying to glue systems together with scripts and workflows, this conversation explains why that approach is already obsolete.

Are coding agents enough, or do enterprises need a completely different class of AI to run their business processes?In this episode of The Agentic Mesh Podcast, we dive into the world of AI agents and the key differences between coding agents and enterprise agents. We explore how coding agents help developers write and iterate code, while enterprise agents tackle complex business processes, governance, and cross-team coordination.We share our experiences from the front lines—what’s working, what’s still early, and how enterprise-grade agent ecosystems are shaping the future of AI in business.Join us as we break down the evolution of coding agents, lessons for enterprises, and the exciting collision of tech and business.

How well does your AI understand your business's decision-making processes? In our latest podcast, we highlight the critical role of policies in context engineering and how they can shape AI responses effectively. Don’t let your AI miss out on this vital information! The context window is one of the most critical and constrained resources in AI. In this episode of the Agentic Mesh Podcast, we explain why context engineering—and treating policies and decision boundaries as first-class context—is essential for building reliable enterprise AI agents.We break down token limits, RAG shortcomings, and the idea of minimum viable context, using real-world examples to show how agents can operate safely, accurately, and at scale.Ideal for anyone working in agentic AI, enterprise AI, knowledge engineering, and AI governance.

In this conversation, Eric and John delve into the critical role of context in AI, particularly focusing on the concept of the 'minimum viable context' and its implications for AI agents. They discuss the challenges of managing context windows, the importance of decision traces and policies, and the need for trust in AI systems. The dialogue emphasizes the necessity of human oversight and the evolving nature of knowledge management in the context of AI.

In this conversation, Eric and John delve into the critical role of context in AI, particularly focusing on the concept of the 'minimum viable context' and its implications for AI agents. They discuss the challenges of managing context windows, the importance of decision traces and policies, and the need for trust in AI systems. The dialogue emphasizes the necessity of human oversight and the evolving nature of knowledge management in the context of AI.

The hard problem isn’t building smarter agents; rather, if we are going to have hundreds, thousands, or maybe millions of agents in each enterprise, then it’s about building the social infrastructure for trust at-scale; it’s about making the entire agent ecosystem and the agents in them transparent, auditable, certifiable – and trusted.In this episode, John Miller and Eric Broda discuss the critical aspects of trust in AI systems, emphasizing the importance of governance, compliance, and certification. They explore how trust in AI agents should be built similarly to trust in human-operated systems, focusing on the need for transparency, accountability, and a robust governance framework. The conversation also delves into the challenges of granting autonomy to AI agents and the necessity of ensuring that these systems can be audited and trusted at scale.

The future is ambient agents: always on, always listening, and always working. The hard part isn’t a prettier UI; it’s identity, state, and audit-ready agent-to-agent handoffs at scale—why A2A, MCP, and an Agentic Mesh exist.In this conversation, Eric and John explore the concept of ambient agents, which are AI agents that operate in the background, always on and listening. They discuss the implications of this new architecture, the challenges of agent communication, and the need for a shift in how we think about agent ecosystems. The conversation highlights the importance of distributed computing, security concerns, and the future of agent collaboration.

Software is getting cheaper to build and insanely faster to ship, shrinking “ten people for a year” into days. However, the real question is this: if the tools are already here and the upside is that big, why are so many companies still stuck watching from the sidelines?In this conversation, Eric and John explore the emerging agent economy, discussing the rapid decline in software costs and delivery times due to advancements in technology. They highlight the potential of software agents to perform tasks traditionally done by humans, the challenges organizations face in adopting these technologies, and the importance of trust and control in the implementation of AI. The discussion also touches on the skills gap in AI, the emerging divide between organizations that can leverage these technologies and those that cannot, and strategic considerations for businesses looking to participate in the agent economy.

Today, far too many AI and agent projects die as science experiments. AI and agent projects live in a lab, impress at demo time, but they never get into production.In this conversation, Eric and John tackle this problem head-on. They discuss the challenges faced by AI projects, during the transition from proof of concept to production. They explore the socio-technical problems that hinder operationalization, the importance of reliability and scale, and the necessity of building trust in AI agents through governance and certification. The discussion emphasizes the need for organizations to learn from past experiences in technology and apply those lessons to successfully deploy AI solutions.

Agentic Mesh is the idea that the real challenge isn’t building a single impressive agent—it’s running an *ecosystem* of them in the real world. In this podcast, we treat agents like infrastructure: how do hundreds or thousands of autonomous agents find each other, coordinate work, and produce consistent outcomes without turning into chaos? Agentic Mesh is the enterprise-grade fabric that makes that possible, with discovery, interoperability, governance, identity, observability, and replay built in—so you can move from isolated demos to a scalable, trustworthy agent ecosystem.In this first episode, John Miller and Eric Broda, both senior data and technology professionals, and active in the agent developer landscape, discuss the emerging field of AI agents and their ecosystems. They explore the importance of understanding agents, their functionalities, and the challenges of implementing them in enterprises. The conversation covers various use cases, the evolution of agents, and the future of agent ecosystems, emphasizing the need for collaboration and scalability in agent technology.