Simulating large-scale enterprises at high resolution lets us measure the ROI of decisions made by AI agents.
Gabriel Tomitsuka
I work on research and engineering at TextQL. I'm interested in pushing the limits of ambitious, high-quality software in the AI age. Most recently, I worked on Argo-Bench, which measures the impact of AI agents' decisions in a high-resolution economic simulation.
In 2023, I joined TextQL as our first employee, to help AI agents understand enterprise-scale data. I've led most of our core platform projects: our ontology and context engine, agent harness, sandboxing technology, and a range of moonshot research efforts. In 2021, I founded Unconvo to make reading social, with shared bookmarks and notes from your friends.
I studied at the Technical University of Munich, where I won the largest hackathon in Germany and learned to love computational theory, data science and machine learning. I started programming at 11, and my first job, at 14, was as a frontend developer at a marketing agency.
LinkedIn / GitHub / Google Scholar / Email
Research
- Orders
- —
- Delivery time
- —
- order to door
- Couriers online
- —
- Courier earnings
- —
- New order
- Long wait
- Delivery margin
- Drag to turn
/