Quantefakt

Writing is free now. Checking is not.

Notes on verification, execution, and what actually happens when you let an agent near a codebase that matters.

  1. The edge is not in the strategy from the video

    Put the rules from trading videos through a backtest, exactly as presented, and a few hold. Most show no edge at all — and the ones that look best on thin stocks lose the most once the cost of trading is real.

    6 min
  2. Two winners on camera, 141,027 trades in the test

    An opening-range fade from a popular trading video, tested exactly as presented across nine years and every stock in our test universe. The video shows two winning trades. The test shows a coin flip.

    5 min
  3. Read-easy beats write-fast

    For twenty years Java lost on being slow to write. A model now writes it as fast as anything else — and what is left to choose a language on is how easily it can be checked. That contest Java wins.

    4 min
  4. We handed the agent the answer. It got it wrong three times.

    Give an AI agent the exact command to run and it still writes its own. It does not copy — it generates from scratch, and a strong enough habit beats the answer in front of it. The concrete case, and why jawata refuses instead of reminding.

    4 min
  5. Parity is not a claim, it is a number

    One time source, a real matching engine instead of a slippage guess, and selection that is bit-exact against the simulator. The same discipline, pointed at trading.

    4 min

RSS →