XSource
We set up an agent to run autoresearch every 3 hours
by Belinda (@belindmo) · 2026-03-14
✓ Opened via the X API on 2026-10-15
Read the original ↗Part 1 · What the source did
An agent wakes on a schedule, reads its own research log, forms a hypothesis, runs a training experiment on a rented GPU, keeps or discards the result, and goes back to sleep.
- Ran an open autoresearch setup every 3 hours with no human in the loop.
- Each run reads the prior log, forms a hypothesis, trains on an A100 and decides keep/discard.
- After 20 experiments the authors said it was too early to tell if it works.
Summary written by Agent Cookbook. See the original for the full context.
Part 2 · Our copy-paste version
Adapted by us from what the source described. Fill in the [BRACKETS].
The prompt
Read research_log.md. Pick ONE hypothesis not yet tested and write it down with the expected result.
Run the experiment with [TRAIN COMMAND] within [N] GPU-minutes.
Append: hypothesis, config diff, metric before/after, KEEP or DISCARD and why.
Stop if spend would exceed [BUDGET] or if 3 runs in a row fail to start.
🛡
Approval ruleBudget cap enforced outside the agent. You review the log before any result is published.
Set it up
- Set a hard spending cap at the GPU provider.
- Seed research_log.md with a baseline.
- Schedule the routine and read the log daily.
Common failures
⚠
What usually goes wrong
- Unattended loops burn money on broken configs.
- The agent 'improves' by overfitting the eval.
Variations
- Run cheaper CPU experiments first.
- Grok Bot routine summarizing the log each morning.