The AI runs my quant research every hour, and one strategy already trades real money
His loop starts the AI once an hour with no human approval, a decision table in code picks one action, and each cycle ends in one pull request that a checking script may merge and deploy. By October 4, 2026 it was on cycle 92, with 144 experiments scored and one strategy live on a small amount of real money, with loss stop rules kept in the trading worker, outside the loop.
How it works
- Every hour a shell script decides whether a cycle should start
- Code, not the model, picks one action from a table of ten
- The AI does that one action and opens one pull request
- A merge script checks the change and deploys the trading worker
- Results are logged and sent to Slack
Why it’s worth knowing
The model does not choose its own work. Plain code picks the one job each hour, the agent does only that, and one of its strategies already trades real money while the stop rules live outside its reach.
Where it’s from
Summarized from a public post on zenn.dev. See the original ↗