Adding a dynamic RAG system to handle build errors#1216
Adding a dynamic RAG system to handle build errors#1216wenta0g wants to merge 43 commits intogoogle:mainfrom
Conversation
|
/gcbrun exp -n wentao -m vertex_ai_gemini-2-5-pro-chat -ag -b quick-test -ns 1 -- --use-error-memory |
|
Requested GKE Job: ofg-pr-1216-wentao JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wentao |
|
/gcbrun exp -n wentao -m vertex_ai_gemini-2-5-pro-chat -ag -b quick-test -rd -ns 1 -- --use-error-memory |
|
Requested GKE Job: ofg-pr-1216-wentao JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wentao |
|
/gcbrun exp -n wentao -m vertex_ai_gemini-2-5-pro-chat -ag -b minor-for-ci -rd -ns 1 -- --use-error-memory |
|
Requested GKE Job: ofg-pr-1216-wentao JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wentao |
|
/gcbrun exp -n wentao -m vertex_ai_gemini-2-5-pro-chat -ag -b quick-test -rd -ns 5 -- --use-error-memory |
|
Requested GKE Job: ofg-pr-1216-wentao JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wentao |
|
/gcbrun exp -n wentao -m vertex_ai_gemini-2-5-pro-chat -ag -- --use-error-memory |
|
Requested GKE Job: ofg-pr-1216-wentao JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wentao |
|
/gcbrun exp -n wentao -m vertex_ai_gemini-2-5-pro-chat -ag -- --use-error-memory |
|
Requested GKE Job: ofg-pr-1216-wentao JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wentao |
|
/gcbrun exp -n wentao -m vertex_ai_gemini-2-5-pro-chat -ag -- --use-error-memory |
|
Requested GKE Job: ofg-pr-1216-wentao JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wentao |
|
/gcbrun exp -n wenqi-include -m vertex_ai_gemini-2-5-pro-chat -ag -b hard -- --use-error-memory --memory-project-filter all |
|
Requested GKE Job: ofg-pr-1216-wenqi-include JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wenqi-include |
|
/gcbrun exp -n wenqi-rq341 -m vertex_ai_gemini-3-1-pro-chat -ll global -ag -b hard -- --use-error-memory --memory-model-filter vertex_ai_gemini-2-5-pro-chat |
|
Requested GKE Job: ofg-pr-1216-wenqi-rq341 JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wenqi-rq341 |
|
/gcbrun exp -n wenqi-rq342 -m vertex_ai_gemini-3-1-pro-chat -ll global -ag -b hard -- --use-error-memory --memory-model-filter vertex_ai_gemini-2-5-pro-chat |
|
Requested GKE Job: ofg-pr-1216-wenqi-rq342 JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wenqi-rq342 |
|
/gcbrun exp -n wenqi-rq343 -m vertex_ai_gemini-3-1-pro-chat -ll global -ag -b hard -- --use-error-memory --memory-model-filter vertex_ai_gemini-2-5-pro-chat |
|
Requested GKE Job: ofg-pr-1216-wenqi-rq343 JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wenqi-rq343 |
|
/gcbrun exp -n wenqi-rq344 -m vertex_ai_gemini-3-1-pro-chat -ll global -ag -b hard -- --use-error-memory --memory-model-filter vertex_ai_gemini-2-5-pro-chat |
|
Requested GKE Job: ofg-pr-1216-wenqi-rq344 JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wenqi-rq344 |
|
/gcbrun exp -n wenqi-rq345 -m vertex_ai_gemini-3-1-pro-chat -ll global -ag -b hard -- --use-error-memory --memory-model-filter vertex_ai_gemini-2-5-pro-chat |
|
Requested GKE Job: ofg-pr-1216-wenqi-rq345 JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wenqi-rq345 |
|
/gcbrun exp -n wenqi-rq346 -m vertex_ai_gemini-3-1-pro-chat -ll global -ag -b hard -- --use-error-memory --memory-model-filter vertex_ai_gemini-2-5-pro-chat |
|
Requested GKE Job: ofg-pr-1216-wenqi-rq346 JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wenqi-rq346 |
|
/gcbrun exp -n wenqi-rq35 -m vertex_ai_gemini-3-1-pro-chat -ll global -ag -b hard -- --use-error-memory --memory-model-filter vertex_ai_gemini-3-1-pro-chat |
|
Requested GKE Job: ofg-pr-1216-wenqi-rq35 JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wenqi-rq35 |
|
/gcbrun exp -n wenqi-rq351 -m vertex_ai_gemini-3-1-pro-chat -ll global -ag -b hard -- --use-error-memory --memory-model-filter vertex_ai_gemini-3-1-pro-chat |
|
Requested GKE Job: ofg-pr-1216-wenqi-rq351 JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wenqi-rq351 |
|
/gcbrun exp -n wenqi-rq352 -m vertex_ai_gemini-3-1-pro-chat -ll global -ag -b hard -- --use-error-memory --memory-model-filter vertex_ai_gemini-3-1-pro-chat |
|
Requested GKE Job: ofg-pr-1216-wenqi-rq352 JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wenqi-rq352 |
|
/gcbrun exp -n wqrq1 -m vertex_ai_gemini-2-5-pro-chat -ag -b comparison-new -- --use-error-memory --memory-project-filter only-current |
|
Requested GKE Job: ofg-pr-1216-wqrq1 JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wqrq1 |
|
/gcbrun exp -n wqrq2 -m vertex_ai_gemini-2-5-pro-chat -ag -b comparison-new -- --use-error-memory --memory-project-filter exclude-current |
|
Requested GKE Job: ofg-pr-1216-wqrq2 JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wqrq2 |
|
/gcbrun exp -n wqrq44 -m vertex_ai_gemini-3-1-pro-chat -ll global -ag -b hard -- --use-error-memory |
|
Requested GKE Job: ofg-pr-1216-wqrq44 JOB: https://console.cloud.google.com/kubernetes/job/us-central1-c/llm-experiment/default/ofg-pr-1216-wqrq44 |
This PR introduces a RAG-based system to handle build errors more effectively. By retrieving previously successful fix snippets ("patch texts") from a database, the agent can reuse past solutions to avoid recurring build issues.
Implementation details:
Under
agent, we add amemory_prototyperagent that extends the general workflow of prototyper, but injects historical fix data into the context when compilation fails.Infrastructure: We add the
memory_helpermodule with utilities to access and connect to cloudSQL instance.Current Status and Purpose: This PR is a draft to verify that the memory_helper utilities can successfully authenticate and connect to the Cloud SQL instance from the CI/CD environment. It is not ready for review.