I’ve seen so many data pros spending hours asking ChatGPT for code and then debugging errors just to analyze a CSV. I have a solution now: just upload your data, ask questions in plain English, and get instant dashboards—no more random code or endless debugging. Powered by SambaNova and E2B for secure code execution, data analysis just got a lot faster.
It is interesting topic, but I’m more curious about the future, how we can combine RAG for data search, and graph relations to map all relevant data. From that point I think LLMs can create code that process the data.
But the most foundational question:
Do we really need BI and dashboards?
Most people has a question, and they need an answer. All other time we spend around the answer is waist.
If we go to the root cause things can be more simple.
I know it is more philosophy than technology, but to decrease AI based consumption we need to answer these questions.
Absolutely, great point! The goal is to move beyond dashboards and just give people direct answers using RAG and LLMs. The demo shows how you can ask a question, search the data, and get the answer—no extra steps. If we keep it simple, we save time and reduce unnecessary AI use. Thanks for sharing your perspective!