Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

how is scheduling conflict resolution handled under the hood — is it hard-coded logic or learned via prompting?



Currently it resolves the conflict via an LLM-call. The plan is to replace that call with a communication routine between the responsible nodes, e.g. also a prompt but instead of directly to OpenAI, to the other node. They should then communicate until resolving the issue and finding a better solution. As soon as we built a personalization and a memory, this should be doable, right?




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: