Environments to act in.
Live tools, real codebases, and stateful workflows. A place for agents to try, fail, and try again.
Intelligence
learns by
doing.
RL environments and high-quality training data at scale for AI labs building models that can do real work.
s0s1s2s3
ObserveActLearn
Our approach
Real work is
the curriculum.
Useful intelligence takes more than the right answer. It takes the ability to act, adapt, and follow through.
We turn complex work into training experiences: environments with persistent state, tasks that unfold over time, and rewards grounded in what actually happened.
Data to learn from.
Tasks and trajectories across coding, science, AI research, and professional work, built around the capabilities your model needs.
Outcomes to verify.
Programmatic checks, deterministic tests, and task-specific rubrics that make progress measurable.
Domains
The world is
more than a prompt.
From long-horizon coding to scientific discovery and AI research, we build environments for work that takes more than one step.
A codebase that keeps evolving.
Agents edit, test, and debug across successive requirements. Each change becomes part of the next challenge.
- Real repositories
- Docker environments
- Programmatic tests
Illustrative taskHarbor format
instruction.mdModernize a Python package without breaking its public interface.
environment/- Python repository, dependencies, and test tooling.
tests/test.sh- Public API compatibility and regression tests.
Research & open work
Make the next
capability measurable.
Benchmark
Multi-step-SWE
One codebase. Successive requirements. A benchmark for coding agents working on software that keeps evolving.
Paper & leaderboard coming soonExplore the benchmarkBuilding the conditions
for better intelligence.
We develop stateful environments, expert-curated tasks, and verifiers for frontier-model post-training. By coupling long-horizon interaction with outcome-based reward signals, we enable reinforcement learning and reproducible evaluation of agent capabilities.
Y Backed by Y Combinator