Above everything I’m genuinely just looking for feedback on my project - I am an parttime indie hacker (student) and I’ve built this over the last few weeks. Currently, I’ve tried tools that allow SQL queries but anything beyond that is impossible which seriously limits how much I could do with data, other than export it to colab and run analysis in python. Ive extended all of this so data retrieval and analysis are tied into one natural language interface.
I’ve add certain features beside prompt engineering that I hope make the product more useful like a debug loop (i.e. feed errors back to the LLM to generate new code), persistent VM connections with high RAM (let’s say I create a linear regression model in one prompt, I can call it in the next prompt, and a micro Python IDE so you can change the generated code itself and run a new version of it.
Any and all feedback + criticism welcome.
No comments yet.