🚨 Only two weeks left until Data Outpost. Nov 4-5 in SFGet your ticket before they're gone
Sep 21, 2026 · Hamilton Ulmer, Till Döhmen, Jacob Matson
Classify text in SQL with prompt_jev(), powered by TypeSafe's Jev. 100,000 rows labeled in 40 seconds for $0.50 at frontier-LLM accuracy. Live on paid plans.
Jul 29, 2026 · Garrett O'Brien, Hamilton Ulmer, Till Döhmen, Jacob Matson
Introducing Guides, MotherDuck's context layer for AI agents. More accurate queries, lower token spend, and distributed context for your entire organization.
Dec 18, 2025 · Till Döhmen
How MotherDuck built a production-ready remote MCP server—from OAuth proxy challenges with Auth0 to tool design patterns that help AI agents query data warehouses effectively. Includes lessons from a hackathon where agents ran 4,000+
29:54
Oct 23, 2024
MotherDuck's Till Döhmen talks at DuckDB Meetup #1 Amsterdam about all the clever ways the AI team is extending SQL to include AI features.
YouTube
AI, ML and LLMs
Meetup
Talk
Oct 17, 2024 · Till Döhmen
We make your database smarter with small language model (and LLM) support in SQL
50:21
Oct 9, 2024
Till and Adithya from MotherDuck discuss DuckDB’s impact on analytics and AI, showcasing its fast, versatile in-process SQL and AI features like text-to-SQL, vector search, and query correction—powerful analytics even on your laptop.
AI, ML and LLMs
Interview
Sep 6, 2024 · Till Döhmen
How to generate a web app dashboard based on your data
Aug 14, 2024 · Till Döhmen
Doing RAG for LLMs or making semantic search results pop? MotherDuck and DuckDB make it easy!
Jul 2, 2024 · Stephanie Wang, Till Döhmen
MotherDuck Founding Engineer Stephanie Wang and AI/ML Lead Till Döhmen recap their highlights and key takeaways from SIGMOD PODS 2024 in Santiago, Chile. Learn more about emerging trends in Text2SQL, Hybrid Resource Allocation, Data Discovery, and more.
Jan 25, 2024 · Till Döhmen, Jordan Tigani
Our first Text2SQL model release!
Jan 3, 2024 · Till Döhmen, Hamilton Ulmer
FixIt will correct mistakes in your SQL queries based on the schema and DuckDB syntax. Based on a large language model (LLM).
