Inspiration
I have always liked network information problems. After a decade in the working world spanning engineering and business roles, I came to the conclusion that spreadsheets were an important, enduring technology, but Excel and Sheets weren't built correctly -- they lack storage, compute, and "cloud" as primitives. Worse, they weren't going to enable future kinds of robust applications (including AI applications). If we didn't take on the challenge, then change wouldn't happen.
What it does
What is Sourcetable? Sourcetable is an AI spreadsheet and data platform that helps you do better, faster work. It's great for analyzing files, cleaning or enriching data, research, data visualizations, reporting, business intelligence, and much more.
Sourcetable is a new kind of spreadsheet with all the same formulas and functions as Excel or Google sheets, as well as data connections to 100+ databases and applications, and helps you pair with the latest AI models to become more productive and data driven.
How I built it
Sourcetable was built to be fast. It was also built to scale.
- AI: LLama 3 (via Groq), Claude, GPT-4o, LiteLLM, custom LLMs
- Frontend: DuckDB, React, ShadCN, AntV / Bizcharts, Plotly, CodeMirror, Hookstate
- Backend: DuckDB, Python, Cassandra, Redis, NGINX, Cloudflare
- Data Eng & Transformations: Fivetran, DBT, Apache Arrow, SQLglot
- Distributed Computing & Scaling: Daft, Ray, Cloud Formation
- Other: Linux Namespaces, Dill (U.Queensland)
Challenges I ran into
Spreadsheets are one of the hardest things you can build. We initially bootstrapped the prototyping and alpha phase, then raised venture capital to recruit a team.
Accomplishments that I'm proud of
Recruiting the team. Everybody is experienced, excellent, and fun to hang around with. It's a good vibe.
What I learned
Start with the smallest product that delivers value and build out from there, while keeping in mind the future architectural vision so you don't make irreversible tradeoffs.
What's next for Sourcetable
Currently, users can do a lot. XLSX analysis and visualizations are a popular use case, as is syncing databases (Postgres, MySQL, etc.) and using the AI to assist with Business-intelligence driven queries. Users can enrich, clean, summarize, translate, and work with larger files than other spreadsheet or chat-driven AI systems, and create world-class visualizations.
The future will be grinding on product quality while expanding our capabilities through AI & applications inside our spreadsheet & data platform.

Log in or sign up for Devpost to join the conversation.