Providers ship silent model changes with no version bump and no changelog. Same 200 OK, same latency — but JSON fidelity slips and outputs drift. Standard monitoring is blind to it. I build the early-warning layer that isn't.
Open-source engine · Apache-2.0 · privacy-preserving by design · you own every result.
For teams shipping on OpenAI, Anthropic, Mistral or any third-party LLM API — where a silent model change means broken JSON, degraded answers, and support tickets before you know why.
Two-person teams pitch with a deck. I pitch with a reproducible, seeded backtest you can re-run yourself.
In the backtest, detection occurred during the 0.8% misrouting window — 38 days before the official postmortem and 19 days before the escalation became visible to users. The methodology is hash-committed, so the next call is a prediction, not a postmortem.
Evidence: live dashboard · open-source engine · dev.to writeup · pip install seismograph-probe · how-to guide
Apache-2.0, open source 193 automated tests, CI-gated CodeQL: 0 open alerts PyPI releases via OIDC Trusted Publishing — no tokens Citable: DOI 10.5281/zenodo.21045517
One email listing the LLM APIs you depend on. No call, no system access, no NDA needed.
Your top-3 silent-drift exposures, ranked, with the exact metrics that would move first. Yours to keep either way.
If the map warrants it: canary probes on your real model tuples, tuned thresholds, alerts wired into Slack or PagerDuty. Fixed price, reviewable deliverable.
Flat pricing, concrete deliverables, fixed timelines. No open-ended “let's chat” retainers to start.
~48 hours · no system access
Zero risk. You leave with a watchlist worth more than the time it took.
~1 week · reviewable deliverable
standing early-warning line
Cheaper than one bad week in production.
Need it built into your stack and handed over working? A full Proof-of-Process Build (in-VPC probe fleet wired to your alerting, runbook, handover) is scoped per engagement — ask in your scan.
Only derived, non-reversible features cross the boundary — token counts, a JSON-valid flag, SHA-256 hashes. Raw text is never stored or transmitted.
Numeric signals carry Laplace DP noise with a tracked epsilon budget. Aggregated, anonymised, defensible.
Probes deploy inside your perimeter. Your data never depends on my infrastructure. Audit-grade evidence export for compliance.
I don't sell you a model, so I have no reason to tell you it's fine. The engine is Apache-2.0 and auditable line by line.
Security posture. Signals are signed with per-org Ed25519 keys. The codebase ships with CodeQL static analysis (0 open alerts), a 193-test CI gate, 2FA-protected accounts, and tokenless PyPI releases via OIDC Trusted Publishing. Reproduce every claim yourself — the backtest is seeded and open.
The Anthropic routing bug degraded output for weeks at 0.8% of traffic while every dashboard stayed green — a seeded backtest flags it 38 days before the postmortem. Your free scan takes one email and ~48 hours.
Silent changes happen behind stable model names — in the Anthropic Aug–Sep 2025 case, a seeded backtest flags it 38 days before the postmortem. Leave your email and you’ll be notified when drift alerts open up for the models on the public board.
No form? Request an alert by mail. One email per confirmed drift event — no newsletter, no sharing.
App-level observability tells you your outputs got worse. It can't tell you whether the model itself changed — and it's often channel-conflicted (it sells to the providers it would have to indict). SEISMOGRAPH watches the provider side, neutrally, across time.
The free scan: your top-3 drift exposures within ~48h. The Baseline: live canary probes, tuned thresholds, a written report, and an alerting plan — a reviewable deliverable in about a week, not an open-ended retainer.
Raw prompts and outputs never leave your perimeter — only DP-noised aggregate features. Probes run in your VPC. You own every artifact. Full technical detail is in the open-source repo.
Because it's falsifiable. The 38-day Claude Sonnet 4 call was made on a seeded, reproducible backtest you can re-run. The methodology is hash-committed up front.
Honest answer: the public network effect (cross-org quorum) compounds with adoption and is early. Today the value is the in-VPC early-warning layer and the neutral monitoring — which stand on their own. I'd rather tell you that than oversell it.
Send me the model APIs you depend on. Within ~48 hours you get your top-3 silent-drift exposures and the metrics that move first — plus a fixed quote if you want the Baseline.
or email me directly: tatyan.radchenko@gmail.com
Early-observer program: the first 10 design partners get the scan free in exchange for honest, publishable feedback.