Agentic AI & MCP
Chatbots were the demo; autonomous agents that take real actions are the product. This is my thinking on the patterns that hold up in production — what to let an agent do on its own, where to keep a human in the loop, and how tools and protocols like MCP make it work at scale.
14 items, 2025 to 2026.
Frameworks
- Agentic API Security The caller is no longer a person. The identity, task-scoping, reversibility gates, and observability an API estate needs once autonomous agents hold the credentials.
- Agent Authority An AI agent needs more than an API token. The framework for assigning a sponsor, mandate, permissions, autonomy and evidence before it can act.
- Agentic AI Patterns From single agents to orchestrated teams — the architecture patterns behind AI that acts, and how to choose between them without over-engineering.
- Model Context Protocol How to assess MCP as an integration contract: where it reduces repeated connector work, and where identity, trust, and operating responsibility remain.
Talks
Writing
- Agentic AI Control Planes: The Guardrails That Make Autonomy Operable A practical architecture for agent guardrails: task boundaries, permissions, approvals, budgets, evaluation, observability and incident response.
- MCP Security and Governance: The Control Map Enterprises Actually Need A practical control map for deploying Model Context Protocol safely: identity, permissions, supply chain, approval gates, logging and operational ownership.
- FinOps for AI: Why Your LLM Bill Is Exploding — and How to Stop It Traditional cloud bills scale with traffic. AI bills scale with autonomy — and an agent stuck in a loop spends like an intern with a corporate credit card. The circuit breakers, caching, and chargeback discipline that keep agentic spend answerable to somebody.
- The Model Context Protocol: A Strategic Reference for Technical Leaders MCP standardises how LLM apps call tools, moving integration from M×N towards M+N. It does not make a connection safe — identity, scopes and review stay yours.
- API Security When the Caller Is an AI Agent Authentication was the part the industry solved. Agents break everything downstream of it — what the caller intends, how far its credential reaches, and how much it can destroy before anyone notices.
- Agentic AI Patterns: From Chatbots to Autonomous Workers A decision guide to agentic architecture patterns — single vs multi-agent, sequential, routing, parallelisation, orchestrator-workers, evaluator-optimiser, and reflection — with trade-offs, failure modes, and the observability you can't ship without.