Analytics failure modes with Jev
On-call already drowns in threshold alerts. Warehouses and anomaly detectors still own the math. Jev judges whether the written incident is actionable given numbers you already computed. It does not detect spikes or query the warehouse.
This unofficial page is the failure modes slice of the analytics alert triage pack. Intent: apply the Jev (TypeSafe System One) decision model to analytics alert triage failure modes. Primary search language: Analytics Jev failure modes. Confirm patterns on docs.typesafe.ai. This site does not sell, issue, or proxy TypeSafe keys. Use a credential you already have from the console or a documented gateway.
Independent angle (cover ≠ clone): Alert-class + actionability on narrative + precomputed bands — not a clone of a BI-alert product or anomaly-SaaS IA.
Analytics use-case context
Analytics alert triage breaks in product-specific ways. This page lists those modes so you can write tests — not a generic “AI can be wrong” essay, and not a rival limitations-page clone.
Hub: Use cases. Compare, when the other tool is the real job: analytics alerts.
Failure Modes inputs
Many failures start as contract violations (distractors, missing alert narrative + metric bands text). Canonical shape:
{
"alert": { "id": "a-991", "title": "Checkout p95 up", "body": "p95 2.1s vs 1.2s baseline; error rate flat." },
"bands": { "p95_ratio_band": "1.5-2x", "error_rate_delta_band": "flat" },
"runbook": { "page": "Page if latency ≥ 1.5× and user-facing checkout, unless error_rate also explains it." }
}Decision signals and actions
- Alerting products own thresholds and mute windows (compare).
jev-1.13is weak at comparing 2.1 vs 1.2 — pass bands (official jaggedness).- A noise Choice is not proof the detector is wrong.
- Paging on a single Noul without severity conjunction creates fatigue.
- No invented MTTA numbers.
HTTP vs application:
| You see | Class | Analytics move |
|---|---|---|
| 401 / 422 / 429 / 529 | Documented HTTP | Fix key/body or back off — errors |
| 200 + flat confidence or Noul ≈ 0.5 | Low confidence | Hold; do not page on-call from Jev alone, and do not auto-suppress user-facing alerts |
| Empty gather | Missing evidence | Skip Jev or ask “is enough information present?” |
Do not treat a Noul of 0.5 as a “medium” analytics alert triage score — it means yes and no are equally likely. Conjunctions stay in your code.
Guardrails and escalation
Fail closed: do not page on-call from Jev alone, and do not auto-suppress user-facing alerts. Schema-safe answers are not factual correctness. TypeSafe’s confidence-gated examples use a lower bar for recoverable reads than for irreversible actions. Those numbers are illustrations. For analytics alert triage, treat page_oncall as the high bar (paging on-call or auto-suppressing a user-facing alert). Tune on labels — see offline evaluation.
Evaluation and rollout notes
Your canary set should include each bullet above.
- Missed pages on planted checkout-down narratives
- False pages / suppress mistakes
- Handoff rate
Pin jev-1.13.0 (the versioned id) after you fit thresholds. jev-latest and the marketing line jev-1.13 can move. Log the response model. TypeSafe’s published list price for jev-1.13 is $0.042 per million input tokens (vendor claim — confirm on the models page); output tokens are free on that same page. Unused distractors still bill as input.
Official Python and JavaScript SDKs read TYPESAFE_API_KEY and retry documented 429/529. This site does not sell, issue, or proxy TypeSafe keys. Use a credential you already have from the console or a documented gateway.
Pack map
| Slice | Page |
|---|---|
| Graph and primitives | decision workflow |
What may enter state |
input contracts |
| What to gather first | evidence collection |
| Atomic rules | policy checks |
| Act / review / abstain | confidence thresholds |
| Reviewer payload | human handoff |
| What to persist | audit trail |
| How it breaks | you are here |
| Labeled replay | evaluation |
| Shadow → canary | production rollout |
FAQ
If the API returns 200, is the decision good? 200 only means the call parsed. Low confidence, Noul ≈ 0.5, or a policy miss are application failures.
Where do official weaknesses live? TypeSafe’s jev-1.13 jaggedness note — distractors, arithmetic, adversarial content. We do not invent more.
Where is the rest of the Analytics pack? Start with Analytics evaluation and Analytics decision workflow. Cluster hub: Use cases.
Can Jev replace the anomaly detector? No. Detectors emit numbers. Jev reads the write-up and your runbook excerpt.
Should we send the whole timeseries?
No. Summarize to bands in code. Distractors hurt jev-1.13 and still bill as input (vendor claim).
What this page does not claim
- Not a warehouse, detector, or APM.
- No MTTA/MTTD benchmarks.
- Not official TypeSafe.
- Official TypeSafe status, or that jev.pro issues API keys.
- That a schema-constrained answer is automatically factually correct.
Disclaimer
This is an independent unofficial site and is not affiliated with TypeSafe AI; official documentation is available at https://docs.typesafe.ai.
Primary documentation: https://docs.typesafe.ai. Hub: Use cases.
Sources
Public TypeSafe or adjacent documentation only. No private claims.