Research · AI
Agents you can hand real work to.
Our AI research asks what it takes for an agent to do useful work on someone's behalf, and to be trusted with it: grounded in facts, checked by default, and supervised where it counts.
Research areas
Agentic software engineering
Agents that plan, write, test and ship software, and know when to stop and ask.
Grounding and verification
Methods that tie every claim and every change back to evidence a person can inspect.
Human supervision
Approval flows, review loops and controls that keep people in charge without slowing them down.
Evaluation
Measuring agents on real tasks, not benchmarks, before they reach anyone's business.
From this research
Blueprint
We find where AI pays off in your business, then automate the workflows that eat your week.
Hands
Personal assistants, agentic staff and team augmentation, run by Coach, your AI chief of staff.
Workshop
Describe the tool you need. Workshop plans it, builds it and hands you working software you own.
Fam
Personal assistants, shared household orchestration and family-office functions, with parents firmly in charge.
Quest
Learn anything, at your own pace, from a first science project to a new career. Spark keeps the curiosity going every day.