Notes on running coding agents like an engineering team.https://agentcrew.to/enhttps://agentcrew.to/blog/done-is-a-claim-until-there-is-evidence/https://agentcrew.to/blog/done-is-a-claim-until-there-is-evidence/Why an agent's completion message is the least reliable artefact in your pipeline, and how evidence-gated completion tied to commits and requirements fixes it.Thu, 17 Sep 2026 00:00:00 GMTverificationagentsprocessHarish Ganapathihttps://agentcrew.to/blog/a-worktree-per-task/https://agentcrew.to/blog/a-worktree-per-task/How isolating every agent task in its own Git worktree makes parallel coding agents safe, why integration should be an explicit merge, and why conflicts must block.Tue, 15 Sep 2026 00:00:00 GMTgitparallelismagentsHarish Ganapathihttps://agentcrew.to/blog/a-prompt-is-not-a-specification/https://agentcrew.to/blog/a-prompt-is-not-a-specification/Coding agents build the wrong thing at full speed unless requirements come first. How stable IDs, acceptance criteria and revisions make plans agents can be held to.Sat, 12 Sep 2026 00:00:00 GMTspecsplanningprocessHarish Ganapathihttps://agentcrew.to/blog/two-approvals-not-zero/https://agentcrew.to/blog/two-approvals-not-zero/Full autonomy is a demo and babysitting is a chore. The right number of human interruptions in an engineering change is two: approve the plan, approve delivery.Tue, 08 Sep 2026 00:00:00 GMTautonomyprocessagentsHarish Ganapathi