PolicyEngine / PolicyEngine/policyengine-api
AI microsimulation summary is flaky
Open
Nobody has claimed this yet.
bug
- Dominant language
- Python
- Stars
- 18
- Forks
- 33
- Avg merge
- 23h 40m
- Merged PRs (30d)
- 22
Description
I thought it had to do with running an AI summary for a policy that had already been run, but when recording found that it also occurred for new policies. Are we logging the Claude calls?
https://www.loom.com/share/4a63c6ee7d9e49039dbd1b929fcefb72?sid=cd742281-a0c5-4233-960b-ea045830f0be
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reproducing the flaky AI microsimulation summary for both previously run and new policies, using the linked Loom for context. Then trace the summary path and Claude-call logging in the API; done means the failure is explained and the relevant calls can be observed reliably.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- ai, backend
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100