Hi HN. This is a project I've been building for the last month.<p>It's a macOS postgres client that uses your OpenAI key to generate and immediately execute queries against your database locally on your machine.<p>I imagine the user being someone familiar with SQL and able to identify errors in the generated code, but regularly needs to write queries to answer relatively simple questions.<p>This is my first software project I'm intending on selling myself and would love feedback on the product and landing page. I'd also love ideas on how to market it to users, as I'm very new to this.<p>In the near future I want to flesh out the features to include:<p>* toggle immediate execution<p>* better export/editing<p>* local model inference<p>* windows & linux support<p>* themes<p>Thanks for trying it out and let me know what you think!
Neat project!<p>Just in case its helpful to you, I maintain pg_query together with my colleagues at pganalyze, which packages the Postgres parser as a library: <a href="https://github.com/pganalyze/libpg_query">https://github.com/pganalyze/libpg_query</a><p>Might be useful to include in your product as a way to run a quick syntax check on the query output by the LLM, without actually connecting to the database and causing an error in the logs.
This looks really cool! Unfortunately, no matter how hard I try, I can't seem to connect to a database. I get the "error returned from database: permission denied for database "postgres" even though I'm specifying all the fields.
Put priority on local models and aims for data analysts. Both SQL syntax and a structure of the database even for big project are simple and users will ask relatively simple questions. I bet you don't even need "large" language models for this.
Nice work! I’m really interested in this but I made the unfortunate mistake of tying all my apps to MS SQL a long time ago. I might still pick this up for future projects.