He helped build the infrastructure that runs the modern internet. First AWS. Then Oracle Cloud Infrastructure. Then Heroku. Then Stripe. Now he's at Docker — and he thinks we're about to need a completely new layer underneath all of it.

When Mark Cavage, President & COO of Docker, joined the company, the question wasn't whether agents were coming. It was whether the infrastructure underneath them was ready.

It wasn't.

Before anyone was talking about agentic workloads in production…Before AI tools started writing, running, and deploying their own code…Before CISOs had a framework for reasoning about autonomous systems…

There was a simple but uncomfortable realization:

Containers were built for immutable, predictable software. Agents want to mutate everything.

In this episode of the Future of Data & AI Podcast, Mark Cavage — President & COO of Docker and one of the founding engineers of Oracle Cloud Infrastructure — joins Raja Iqbal for a candid conversation about what the agentic era actually demands from infrastructure.

Mark has spent over two decades building the systems that power modern cloud. Through Docker, he's now working on the sandbox layer that lets enterprises deploy agents at scale — without handing over control to a system nobody fully understands yet.

This conversation goes beyond the hype.

What You'll Discover:

• Why containers alone aren't enough for the agentic era.Containers were built for immutable software. Agents mutate, write, and act — and Mark explains exactly what breaks, and what Docker built to fix it.
• What YOLO mode actually means — and why it matters.Agents running without a human in the loop sounds reckless. Mark explains why that's actually the goal, and how the micro VM sandbox makes it safe enough for enterprise.
• The 1000x risk surface no one is talking about.Every AI-generated pull request, every "authored by Claude" commit, every autonomously deployed dependency is stacking security debt. Mark breaks down what that means for your CISO.
• Trusted MCP servers and Docker Hardened Images.What they are, why they exist, and why supply chain security for AI tools is about to become one of the most important conversations in enterprise tech.
• Mark's bets for the next 12 months.CFOs demanding ROI on token spend, the open source project that no one is talking about, and what the future of Agentic AI looks like.

This episode is for:

• ML engineers and DevOps teams building with agents
• CISOs and security leaders managing a 1000x larger risk surface
• Platform and infrastructure leads evaluating MCP servers and supply chain security
• CTOs and engineering leaders figuring out what "agentic" actually means for their org
• Founders deciding where the next infrastructure layer gets built

This isn't a conversation about demos or roadmaps.

It's about the infrastructure that agents actually need to run safely, reliably, and at scale — and whether the industry is building it fast enough.

If you're deploying agents in production, managing the security conversation, or trying to understand where Docker fits in the agentic stack… this episode is worth your time.

Explore all recordings

🎙️ Future of Data and AI Podcast: Episode 09 with João Moura

He just wanted to automate his own work, but it turned into an AI platform that is now used by top enterprises. How?

When João Moura, founder and CEO of CrewAI, started building AI agents; it wasn’t to launch a business, he just wanted to make his job easier. Starting at Clearbit, he built a set of agents to automate content creation to post more on LinkedIn… and ended up generating up to 600 inbound leads a day.

Instead of relying on a single AI tool, João’s multi-agent systems divide big tasks into smaller ones, letting specialized "crew" of agents collaborate to get work done.

That experiment evolved into CrewAI — a platform where teams can manage, deploy, and track crews of AI Agents handling routine workflows, all from a single dashboard. Today, CrewAI is redefining how the world’s largest organizations leverage multi-agent systems, from Oracle to the U.S. Department of Defense.

Before AI agents became the hottest trend in enterprise tech…
Before every company had 50+ use cases mapped out…
Before “autonomous workflows” became a boardroom priority…

There was a simple but uncomfortable realization:

Building agents is easy. Deploying them is hard.

In this episode of the Future of Data & AI Podcast, João Moura — CEO of CrewAI and former Director of AI Engineering at Clearbit — joins us for a deep dive into the real state of agentic AI.

João has spent over 20 years in software engineering and AI leadership, and through CrewAI, he’s working with major enterprises — including government agencies and global corporations — to move AI agents from proof-of-concept to production.

This conversation goes far beyond hype.

• Why building agents has zero value — unless they see the light of day.
  João explains why most AI projects stall after POCs and what separates experiments from real business impact.
• The biggest misconception about agentic AI.
  AI agents aren’t magical. They require strong engineering discipline, governance, and architecture — sometimes more software engineering than AI.
• From guardrails to trust.
  How enterprises handle non-determinism, hallucinations, prompt injection, and MCP security — using layered guardrails, LLM-as-a-judge, role-based access control, and observability.
• The real bottleneck isn’t intelligence — it’s deployment.
  Models are already good enough. The hard part is compliance, traceability, monitoring, and integrating with existing enterprise systems.
• Why simplicity beats over-engineering.
  João shares a powerful insight: complexity never disappears — you just choose where to put it. Senior engineers understand this. Many teams don’t.
• How enterprises mature with AI.
  Organizations typically start with cost savings, move to revenue generation, and eventually unlock innovation they hadn’t imagined before.
• Open source vs. enterprise AI.
  Why CrewAI chose an open-source-first strategy — and how the business model works when moving from prototyping to production-grade orchestration.
• Advice for AI founders.
  João speaks candidly about entrepreneurship in the AI era — the hype, the competition, investor psychology, and why this may be a once-in-a-generation opportunity.

This episode is for:

• AI engineers building multi-agent systems

• CIOs and enterprise leaders deploying autonomous workflows

• Founders launching AI startups

• Teams struggling to move from POC to production

• Anyone thinking seriously about trust, governance, and scale in AI

This isn’t a conversation about shiny demos.

It’s about building AI systems that run reliably, securely, and at scale — in the real world.

If you’re working on multi-agent systems, autonomous workflows, enterprise AI infrastructure, or AI entrepreneurship… this episode will change how you think about shipping AI.

🎙️ Future of Data and AI Podcast: Episode 08 with Emil EifremBefore graph databases were everywhere…Before knowledge graphs became essential to AI…Before LLMs, embeddings, and RAG entered the conversation…There was one simple, stubborn idea: data makes more sense when you understand the relationships.In this engaging and insightful episode, Emil Eifrem (Co-Founder & CEO | Neo4j) shares the story behind building Neo4j — the graph database platform that quietly reshaped how the world models and reasons over data. Emil takes us through the early days when graph databases felt risky, misunderstood, and years ahead of the market, and explains why connected data mirrors human thinking better than tables, rows, and columns ever could.What you’ll discover:🔹Why relationships became the foundation of Neo4j — and how treating them as first-class citizens changed everything.🔹How property graph models preserve context, making complex systems easier to reason about and analyze.🔹Unexpected ways graph databases are powering fraud detection, recommendations, cybersecurity, and enterprise knowledge.🔹Why knowledge graphs quietly became essential to modern AI — long before most people noticed.🔹How graphs and LLMs work together to ground AI systems in structure, meaning, and explainability.🔹Insights on building deep infrastructure technology with patience, conviction, and long-term vision.This is more than a conversation about AI, graphs, or databases. It’s a look at how intelligence — human or artificial — depends on connections, context, and understanding.📌 If you’ve ever wondered:- Why AI sometimes feels confident but wrong- How machines can reason instead of just predict- Why some technologies take years before the world catches up…this episode will change how you think about data, AI, and connected systems.🔹A must-listen for: data scientists, AI practitioners, knowledge graph enthusiasts, graph database users, enterprise architects, AI researchers, and anyone curious about how connected data powers modern AI.Prepare to walk away inspired.Keywords: Emil Eifrem, Neo4j, graph databases, knowledge graphs, property graph model, AI reasoning, connected data, enterprise AI, graph analytics, fraud detection AI, AI infrastructure, LLM applications, RAG, embeddings, AI explainability, building data platforms, AI strategy, human-like reasoning, data science insights, graph AI, Neo4j interview, AI podcast, Future of Data and AI PodcastFor more episodes: https://www.youtube.com/playlist?list=PL8eNk_zTBST_jMlmiokwBVfS_BqbAt0z2For highlights, check out: https://www.youtube.com/playlist?list=PL8eNk_zTBST-YYNgPcw3rO9Tvn7fEjR4AVisit our podcast page for more info: https://datasciencedojo.com/podcast/