1
Curious builder0 XP earned · 300 to level 2
0 daysFinish a lesson to begin
Badge collection0 of 6 unlocked
51 small wins to finish your pathNext question →
What non-functional requirements are specific to AI systems?
30-second answerSay your answer out loud first, then reveal.
| Requirement | Example target | Why it's AI-specific |
|---|---|---|
| Quality | ≥ 90% correct on golden set; < 2% unsupported claims | Outputs are probabilistic; need evals |
| Latency | TTFT < 1s, full answer < 6s (chat); < 300ms (autocomplete) | Token generation is slow and variable |
| Cost | < ₹2 per conversation; < $0.01 per document | Token costs scale with usage and context |
| Safety | Zero PII leakage; harmful content blocked | Models can be manipulated |
| Freshness | New docs searchable within 1 hour | Model knowledge is frozen |
| Reliability | Graceful fallback if a provider fails | External model dependencies |
| Explainability | Citations; decision logs | Trust, regulation |
| Determinism | Same input, consistent decision (for workflows) | Sampling variance |
Trade-off triangle: quality ↔ latency ↔ cost. A bigger model, more context or more agent steps raises quality but increases latency and cost. Every design decision should say which corner it favours.
Clarifying questions to ask
- What does a "good" output look like? Who judges it?
- What's the cost of a wrong answer (annoying vs dangerous)?
- Interactive or batch? Expected volume and peak?
- What data can leave our network?
Related
Slow is fine. Stopping is the only problem.