Starting clean when a path goes nowhere
The first repo’s best learned bot won 12 of 140 games against the rules bot it was meant to replace. On 31 Oct I had everything saved, deleted it, and started a new repo with only what had proved its worth.
What the new repo took
| Carried over | Left behind |
|---|---|
| The official game docs, unchanged | The orchestrator, its heartbeat and merge queue |
| The official engine and its native build | The gates and the parameter tuner |
| 70 maps, 25 edge-case maps | The learned bot and its training code |
| v1 and every fifth version to v50, as opponents | Old commits, except as reference |
| The map, replay and fuel tools | Every dashboard page but compute |
Proofs and plans: four days, then deleted
The new repo began with a plan I wanted proved: every step citing a register of axioms, backed by a Lean proof, reviewed until reviewers found nothing. Four days in, no version beat v50. Rigour about the plan was not progress on the bot, so I switched to building the bot.
- A register of axiomsAxioms, facts and decisions every step had to cite. 63 entries, then 85.
- Lean proofs and nonstop review“If we can formalize this somehow through some logic, maybe in Lean.”
- A second botv50 with every assumption written down as a cited “belief”.
- I see the plan has no end20 prerequisites before the loop could start. “It seems like it's just a never-ending path.”
- A new bot from first principlesThe lineage that shipped. Its first commit came 53 minutes later.
- The proofs and plans are deletedThe Lean proofs, the plan, the axioms and the second bot.
What the plan said of its own main theorem
today ε_cov has no useful bound, so x = 1 and the theorem holds trivially
A proof can be correct and still promise nothing. I had asked for the heaviest of this machinery myself, and I was the one who had it deleted.
Review rounds: 37, then 3, then none
The rule was fresh reviewers who did none of the work, round after round, until a round found nothing. Every round cost a fresh set of agents.
Highest review round reached, 31 Oct to 5 Nov
Agent threads started on 3 Nov, by job
- WrappersHaiku, each running one review command
- Workflow stepsOpus: fix, triage, record
- Other Opus threadshigh or extra-high effort
- ReviewersGPT 6.1
How I cut the rule back
- Nonstop1 Nov
- One reviewer, every lens3 Nov
- At most 3 rounds4 Nov
- None by default5 Nov
- Only when I ask6 Nov
One night, 353,000 lines
On 6 Nov I asked for everything legacy to go. Most of it went between 02:00 and 05:00.
Lines deleted, by what they were
The mutation tests, planted bugs that check each test catches what it claims, show the trade most clearly.
What the mutation tests cost, and bought
The bot that shipped grew up in the lighter repo: its rating went from 2,244 on 4 Nov to 2,660 at the ladder freeze.
Instructions grow; I cut them back
Most rules came from one incident each, so each repo’s main instruction file grew until agents got lost in it. Three times I had it cut back.
Root instruction file, in KB
“Leave no references to legacy stuff.”
“Set up the repo so that AGENTS.md is clean and clear.”
Keep the bench that pays
From 7 Nov agents built a scored bench for everything they wanted to climb. The ones that sped up tools paid; no bench of the bot’s play was ever shown to predict strength. On 11 Nov I kept one.
The last loop goes too
My last orchestration loop had its own tools, dashboard and ledger of bets. When it started to get in the agents’ way, I had all of it deleted at 23:58 on 12 Nov.
“Delete everything related to the loop.”