
Founder, CEO
Dr. Qingchen Wang
- VC‑backed founder
- Kaggle Grandmaster
- Former University of Hong Kong professor
AsapiThe AI Agent TrialsWe turn real jobs into reinforcement‑learning environments. An agent works through a whole task in a simulated business or household, with files, tools and sometimes people to talk to, and a grader turns what it hands in into a reward. We've built 80 so far, for example:
Many are jobs someone might hand to an assistant, asked the way they'd ask. In most, what the job really needs sits in files, records or people's heads, so the agent has to find it before it acts.
In some environments the agent has to ask people for what it needs, and they answer only what they're asked, as busy colleagues do. In others, the work runs over simulated days, weeks or months, and the agent keeps only its own files between sessions.
In most environments, each of dozens of requirements earns part of the reward, so an agent earns credit for what it gets right. Some also have make‑or‑break requirements, and in some a right decision earns full credit only if the agent first found the fact that decides it.
We tested 28 models on these environments, each given only the request, with no hints and no rules spelled out.
Each dot is one attempt, and the line marks a model's median. The strongest models typically earn around four‑fifths of the reward, and fully solve about one attempt in ten.
The six models we've tested on at least 40 environments.
Share of attempts that made each mistake:
Never checked a rumour that one family wasn't coming, and booked too few seats
Left the broker's client IDs in the app's code or tests
Didn't page a doctor in time about a wrongly charted dose
Bought a new cabinet when a cheaper repair would fix the old one
Each environment targets several. A capability's page lists its environments and how each model did.
Across the environments, code decides most of the reward, by reading the final state of the simulated world and the files the agent leaves behind. Where a judgment is needed, such as whether a letter says the right thing, AI judges answer fixed yes‑or‑no questions.
The three samples you can download show how we package an environment. The grader and its answer key sit outside the agent's reach, a reference solution earns the whole reward, an empty attempt earns nothing, and any judge or model‑voiced person runs on whichever model you choose.
Each with its files, tools and grader, and its simulated people and clock where it has them.
New, unpublished instances of an environment, as many as you need, so you can keep some back for evaluation.
Environments for your own domain, tools and users, graded the same way.
The attempts we made on each environment, with the grader's report on each scored one, to compare yours against.
The samples come as Harbor tasks, an open format for agent tasks, and we package other environments the same way on request.
Exclusive or non‑exclusive, at launch prices.
We trained a 14B DeepSeek model on a simplified version of one environment. On instances it hadn't seen, its score went from 20 to 97 out of 100.
Machine learning researchers with a combined 60+ years of experience.

Founder, CEO

Founding advisor












Three environments are packaged as samples you can download, with their graders and reference solutions. Ask for them with your work email, or book a call about the rest of the catalogue.