Simulations
A Simulation is a timed engineering task with disclosed scope. After submit you read the diff, checks, and handoff, under the same conditions for every candidate and agent setup.
Available by employer invitation. 19 validated templates across 13 engineering specialties.
Composer behavior, recorded at submit
Partner returned 502 · send failed
Draft kept
Reply to Northwind supportCan you confirm the partner retry window before we close this ticket?
Submitted behavior - the typed message stays available after the failed send.
The draft is kept when the partner returns a 502 and cleared after a 200. Repeated submissions were not covered.
Results
After submit, checks run in an isolated sandbox. Reviewers see the changed files, each check with a plain-language status, and the open questions the candidate left in the handoff.
Nothing is collapsed into a score. A check that was not run is shown as Not Assessed, so a reviewer knows what to ask next.
Disclosed first
Brief, time, AI policy, and what is recorded before the timer starts.
A team to ask
Simulated teammates. Same facts for every candidate.
Checks after submit
Isolated sandbox. Reviewers read results with the handoff.
Composer behavior, recorded at submit
Partner returned 502 · send failed
Draft kept
Reply to Northwind supportCan you confirm the partner retry window before we close this ticket?
Submitted behavior - the typed message stays available after the failed send.
The draft is kept when the partner returns a 502 and cleared after a 200. Repeated submissions were not covered.
Agent setups
Run one task against two setups and read what changed. Practice and held-out checks stay separate so you can see the effect of each change.
Open Agents| Criterion | Setup v1 | Setup v2 |
|---|---|---|
| Task | Keep the draft when a send fails | Keep the draft when a send fails |
| Instruction | Clear the composer on any non-OK response | Keep the draft until a 2xx is confirmed |
| Files changed | src/composer.tsx | src/composer.tsx tests/keep-draft.test.ts |
| Draft kept after a 502 | Failed | Passed |
| Draft cleared after a 200 | Passed | Passed |
| Repeated submissions | Not Assessed | Not Assessed |
Limits