Hey everyone - Mathias here, one of the founders. I kept watching data scientists on my team spend hours on work that followed the same pattern every time. Load data, explore it, clean it, build a model, make some plots, write up findings. The steps were predictable but the execution still took forever because of all the little things — debugging imports, fixing plot formatting, restarting sessions, re-running cells in the right order. I tried using AI assistants alongside notebooks but it always felt disconnected. You'd get a code suggestion, paste it in, run it, get an error, go back to the chat, paste the traceback, get a fix, paste that in. It was just copying and pasting between two windows. So we started building something where the AI actually lives inside the notebook. It writes cells, runs them, sees the output, hits errors, fixes them, and keeps going. You end up with a real notebook you can read through, edit, and share — not a chat log. And because everything runs on a cloud runtime, you can kick off a long analysis, close your laptop, and come back later to a finished notebook. The agent keeps working whether you're watching or not. It took a lot longer than expected to get right. Getting an agent to reliably work through a multi-step analysis without going off the rails is a hard problem and we're still improving it. But it's at the point now where it genuinely saves time on real work, and we want more people using it so we can keep making it better. If you work with notebooks regularly I'd love to hear what you think. Happy to answer any questions here.

Avenlo

Accelerate Data Science, Intelligently.

Accelerate Data Science, Intelligently.