widgrensit https://widgrensit.com/ The agent fleet, in practice https://widgrensit.com/blog/the-fleet-in-practice/ 2026-06-26T00:00:00Z The principles are easy to write down. This is what running a fleet of coding agents actually looks like day to day - the handoffs, the gates I stop at, and the failure mode that taught me to never trust a single agent's finding. AI shrinks the team, not the manager's job https://widgrensit.com/blog/smaller-teams-two-leaders/ 2026-06-23T00:00:00Z A bet on how companies will need to organize to drive products forward. AI lets smaller groups do more, but the answer isn't fewer managers. It's separating two leadership jobs big orgs blur together - the PO drives what, the EM grows who - so small teams own the how and never end up stranded. Intent over mandates - leading the move from legacy to modern tooling https://widgrensit.com/blog/intent-not-mandates/ 2026-06-18T00:00:00Z You can mandate a migration and get malicious compliance, or you can give intent and get teams that own the destination. Build a paved road, then prove it shipped value with DORA. Why I reach for the BEAM in regulated systems https://widgrensit.com/blog/why-the-beam-for-regulated-systems/ 2026-06-12T00:00:00Z The properties auditors care about - isolation, observability, graceful degradation - are the same ones OTP was built around 30 years ago. Treating coding agents as a fleet, not a chatbot https://widgrensit.com/blog/agents-as-a-fleet/ 2026-06-05T00:00:00Z The leap in productivity isn't a smarter single agent. It's structuring many narrow agents with clear contracts, gates, and verification.