Special: How Tank Warfare Tests AI Coding Agents

From Ringside & Error Bars: The Battle Report Podcast · 24:30

Show by battle.training

The hosts dig into Bench v1.1, the benchmark built on battle.training: what it measures (a coding agent writing, testing and rewriting a tank controller under a fixed budget), how a controller is scored against twelve frozen opponents on 32 private maps, how the maps are sealed by hash and why every replay is byte-identical. Then the Controlled Edition pilot, where Gemini 4 Argon in Antigravity, GPT-6.1 Terra in Codex and Claude Fable 5.1 in Claude Code each got one prompt and two hours, the bug that gave them unequal time, and what the strongest controllers do on the battlefield. Two corrections to the audio: the Gemini runs uploaded two versions on average, not three; and since the recording, the report carries an erratum: the Codex runs' reasoning effort was never pinned (one ran at xhigh, four at the CLI's default).Chapters:0:00 One submarine, total silence1:18 Beyond pass-or-fail tests3:15 The sandbox and the draft5:06 Terrain, bridges, spotters and sonar7:49 What Bench v1.1 measures9:38 Sealed maps and the overfitting question11:38 Byte-identical replays13:32 The Controlled Edition pilot14:54 The resume-cap bug16:46 The top score and the scientific method18:58 Damage, not volume of fire22:01 Strategy without doctrineAn AI-generated audio discussion, made with Google NotebookLM from the Bench v1.1 report, the arena's documentation and the Battle Report's coverage. The voices are NotebookLM's hosts; the show notes were written by AI from a transcript of the audio and checked against the report.Show notes: https://battle.training/report/podcast/how-tank-warfare-tests-ai-coding-agents

More episodes of Ringside & Error Bars: The Battle Report Podcast

More episodes

PodSearch is a free, independent podcast search engine. Read about PodSearch, the privacy policy or contact us.