Skip to content

Speaker. Builder.

Agentic AI, in production, without getting burned

Thirty years across operations, cloud and platform engineering, from CTO to individual contributor, and now AI Business Stream Lead for Leaders and Organizations at Accenture LearnVantage. I work with executives on the part of agentic AI that decides whether it survives contact with production: governance that infrastructure enforces, not the model. That is the premise of Agentic Covenants, and it is what the MCP Dev Summit Toronto keynote covers, deploying the Model Context Protocol across a workforce of about 750,000 people. Earlier stops: Red Hat, ThoughtWorks, Amazon Web Services, and The Performant Professionals, the advisory practice I founded. The writing here is the working notes.

GitHubLinkedInBlueskySessionize

Michael Rishi Forrester, head and shoulders, in an open office with windows behind him

Next talk

MCP Dev Summit Toronto 2026 (keynote)

6 October 2026

Latest post

Deep Dive: I Let Claude Build the Entire IDP. Here's What That Actually Looks Like.

Latest recording

Ignite Talk - The Day Claude Code Deleted My Cluster - A Cautionary Tale About AI Guardrails

15 May 2026

Agentic Covenants

Guardrails enforced by the AI tool are bypassable. Guardrails enforced by infrastructure are not.Read the covenants

Upcoming

Writing

All 126

Listen

The AI Inevitable, with Whitney Lee. Evidence-based answers to the questions everyone is asking about AI. One falsifiable question per episode, reviewed against the papers.

Play