The team is testing an internal process that utilizes Jev to identify specific rule failures in code, which are then verified by an LLM. This approach appears effective and may reduce costs by addressing simpler issues before conducting more thorough reviews.
added by @dkubb
We’ve been testing something internal that does a lot of the same thing. One difference is we’re using Jev to look at the code and flag specific rule failures and then we hand those to an LLM to verify.
It works pretty well and seems like it can be a bit cheaper to find the low hanging fruit prior to more in-depth passes.
View similar
Ranked from stored criteria vectors. No live classification on this page.
This approach involves using Codex to create APIs for web applications, which are then connected to ChatGPT via custom GPT Actions. With GitHub access to the project code, ChatGPT can provide insights about the app's architecture and behavior.
The author shares insights on integrating a tool for spotting problematic database migrations in their CI/CD pipeline and experimenting with another tool to filter unnecessary jobs based on git diffs.
The Spectra system utilizes Deepgram Flux for speech-to-text functionality, providing end-of-turn reports with probabilities. This setup ensures high accuracy and quick responsiveness for agent interactions.
This project utilizes ChatGPT with its own cloud VM and includes specific instructions for code editing and management. It emphasizes the importance of following project rules and saving progress checkpoints.