Core services
Web app
Responding normally
ok
Database
Connected
ok
Lucida AI (any task)
No recent traffic — status will populate after the next scored candidate.
ok
LLM tasks (last 24h)
| Task | Last success | Last latency | State |
|---|---|---|---|
| clarify_jd | never (no traffic since restart) | — | unknown |
| generate_questions | never (no traffic since restart) | — | unknown |
| generate_rubric | never (no traffic since restart) | — | unknown |
| interviewer_briefing | never (no traffic since restart) | — | unknown |
| score_candidate | never (no traffic since restart) | — | unknown |
| suggest_channels | never (no traffic since restart) | — | unknown |
| suggest_hiring_angles | never (no traffic since restart) | — | unknown |
| suggest_knockouts | never (no traffic since restart) | — | unknown |
"Stale" means the task hasn't succeeded in the last 24h. Often that's just low traffic, not an outage.
During a Lucida AI outage, scoring requests queue at status=SUBMITTED and are auto-resumed
when service recovers — no candidate loses their application.
If something is degraded
- Database degraded — admin actions may fail. Try again in a few minutes; if it persists, email hello@lucida.team.
- LLM degraded — new applications are accepted and queued. Scoring resumes automatically when the upstream API recovers. No data is lost.
- Both degraded — apply forms may show errors. We watch this surface live and will post an update here. Razorpay billing and existing data remain unaffected.