Problems / AI products, medium
A voice agent transcribes speech (100–150 ms), asks a model (first token: 250 ms usually, 2.4 s for the slowest 1%) and speaks the answer. The call to the model waits up to 8 s and has nowhere else to go. Callers notice any silence over about a second.
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 ≥ 99%, time to first token p99 ≤ 1.2 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.
60 requests a second for 30 seconds: 100% request (reads).
What each component does in this design, and what calls it. The simulator mimics each one from these settings.
| Component | Its job here | Settings |
|---|---|---|
| WebSocket gatewayWebSocket gateway | Traffic enters here.Calls Load balancer.WebSockets hold a connection per client, so memory grows with connected users, not requests. | 50000 slots |
| Load balancerLoad balancer | WebSocket gateway forwards to it.Calls AI app.Layer 4 balancers route by IP and port; layer 7 balancers read HTTP and can route by path or header. | – |
| AI appApp service | Load balancer forwards to it.Calls Speech to text (streaming), Model, Text to speech.Little's law: throughput ≈ concurrency ÷ latency. 200 slots at 50 ms serve about 4,000 req/s. | 3 replicas × 500 slots · timeout 8.0 s |
| Speech to text (streaming)Speech to text | AI app calls it.Streaming speech recognition returns partial words while the user is still talking. | 2000 slots · 6,000 calls/min |
| ModelLLM endpoint | AI app calls it.Latency has two parts: time to the first token, then a steady stream of tokens per second. | 2000 streams |
| Text to speechText to speech | AI app calls it.Speech synthesis can start playing before the whole sentence is generated. | 2000 slots · 6,000 calls/min |