I kept hitting the same wall with "chat with your data" tools: an AI gives you a confident number, and you have no idea if it's real or hallucinated. Curator fixes that by refusing to bluff. Ask a question in plain English and it writes real Python (pandas, numpy, scipy), runs it on your data, and hands you the answer with the exact code that produced it so you can check the math yourself. It doesn't guess your average, or your clusters; it computes them. Two things fall out of doing it this way: 🔒 It runs in your browser, so your data never leaves your machine. 🧠 It learns — your analyses become reusable building blocks, so it gets sharper the more you use it. It's early and I'm building it mostly solo. I'd love for you to throw a real, messy file at it and tell me where it breaks. What would make this genuinely useful for your data?

AskCurator

Chat with your data. See the code. Teach it your workflows.

Chat with your data. See the code. Teach it your workflows.