Field notes on agentic engineering, kernel sandboxes and human-in-the-loop AI for coding agents. https://tamer.ai/en/blog/ 2026-06-15T00:00:00Z tamer team https://tamer.ai BSL 1.1 — Apache 2.0 after 4 years https://tamer.ai/en/blog/2026-06-15-beyond-the-context-window-surgical-investigation/ 2026-06-15T00:00:00Z 2026-06-15T00:00:00Z fred

Dumping your entire repo into a 1M token context window is a recipe for drift. Learn why Tamer treats the filesystem as the source of truth, using surgical investigation to maintain high-signal supervision.

https://tamer.ai/en/blog/2026-06-10-tamer-is-not-a-claude-alternative/ 2026-06-10T00:00:00Z 2026-06-10T00:00:00Z fred

You keep your Claude Max / Gemini Advanced subscription. Tamer orchestrates the agents you already pay for — sharper specs, drift detection, and an audit trail SOC teams can actually use.

https://tamer.ai/en/blog/the-7-section-self-critique-gate/ 2026-05-31T00:00:00Z 2026-05-31T00:00:00Z tamer team

To move fast, you must be your own harshest critic. Inside the 7-section block that tamer enforces on every single Work Item [DONE] event.

https://tamer.ai/en/blog/tamer-multi-agent-supervision-patterns/ 2026-05-31T00:00:00Z 2026-05-31T00:00:00Z tamer team

Three patterns that turned a chaotic swarm of agents into a reliable production pipeline.

https://tamer.ai/en/blog/tamer-copilot-field-report/ 2026-05-31T00:00:00Z 2026-05-31T00:00:00Z tamer team

We tested the integration of GitHub Copilot CLI as a tamer worker. Here is why, despite the model's power, automation still stumbles on implementation details.

https://tamer.ai/en/blog/multi-agent-stale-state/ 2026-05-31T00:00:00Z 2026-05-31T00:00:00Z tamer team

How a 16-day bug in tamer's worker discovery taught us that in multi-agent systems, the filesystem is a liar. The move to server-side live truth.

https://tamer.ai/en/blog/2-day-cadence-slipped/ 2026-05-31T00:00:00Z 2026-05-31T00:00:00Z tamer team

The irony of automation: tamer promises a breakneck delivery cadence, but we missed our own goal. Here is why, and what we changed.

https://tamer.ai/en/blog/2026-05-17-terminal-transparency-cost-control/ 2026-05-17T00:00:00Z 2026-05-17T00:00:00Z tamer team

AI shouldn't be a black box. Discover why tamer prioritizes Standard Input (STDIN) and respects your existing subscriptions.

https://tamer.ai/en/blog/2026-05-13-building-tamer-spec-completeness-v2-five-axis-study/ 2026-05-13T00:00:00Z 2026-05-13T00:00:00Z tamer team

Five axes, five iterations each. How we turned 'is this spec good enough?' from a vibe into a checklist that a Master Agent can refuse a Work Item against.

https://tamer.ai/en/blog/2026-05-09-anatomy-of-a-done-event-that-never-arrived/ 2026-05-09T00:00:00Z 2026-05-09T00:00:00Z tamer team

Four levels of root cause for why an agent that finishes its work stays invisible.

https://tamer.ai/en/blog/2026-05-07-why-your-ai-agent-is-silently-degrading-your-codebase/ 2026-05-07T00:00:00Z 2026-05-07T00:00:00Z tamer team

75% of AI agents see their regression rate grow over time. Here's how to avoid it.

https://tamer.ai/en/blog/2026-05-05-spec-is-the-bottleneck/ 2026-05-05T00:00:00Z 2026-05-05T00:00:00Z tamer team

Once model capability passes a threshold, code quality depends on the completeness of the specification — not the intelligence of the model. A field note from tamer's design log.