Problems / Agents and tools, medium
A coding-agent platform runs 7 tasks a second with 3 sandbox runs each, about 3.5 s per run, on a provider that allows 100 sandboxes at once. On 30% of runs the sandbox asks a person to approve a command and stays reserved (and billed) while it waits, about 12 s. Times are scaled: a second here stands for a minute.
Unlock it once for 15 credits; everything inside it is free (runs, changes, Submit, results, hints). Sign in with Google to start; new accounts get free credits.
Success ≥ 95% of tasks, p99 ≤ 90 s.
Like a coding problem, a solution has to pass every case, not just the one it was built for. Submit in the lab runs all 6; 2 of them are hidden until then.
7 requests a second for 45 seconds: 100% task (writes).
What each component does in this design, and what calls it. The simulator mimics each one from these settings.
| Component | Its job here | Settings |
|---|---|---|
| Load balancerLoad balancer | Traffic enters here.Calls Agent API.Layer 4 balancers route by IP and port; layer 7 balancers read HTTP and can route by path or header. | – |
| Agent APIApp service | Load balancer forwards to it.Calls Agent orchestrator.Little's law: throughput ≈ concurrency ÷ latency. 200 slots at 50 ms serve about 4,000 req/s. | 3 replicas × 500 slots · timeout 180 s |
| Agent orchestratorAgent orchestrator | Agent API calls it.Calls Model, Code sandboxes.An agent loops: think, call a tool, read the result, repeat until done. | 400 slots · 3 steps |
| Human approvalHuman approval | Code sandboxes calls it on 30% of requests.Human-in-the-loop review holds risky actions until a person approves them. | 1000 slots |
| ModelLLM endpoint | Agent orchestrator calls it.Latency has two parts: time to the first token, then a steady stream of tokens per second. | 4000 streams |
| Code sandboxesCode sandbox | Agent orchestrator calls it.Calls Human approval.Sandboxes run model-written code in an isolated container with limited network and time. | 100 slots |