Genome · Behavioural anomaly engine
What's out-of-pattern — and why.
Genome learns your tenant's normal patterns (transactions, agent actions, login behaviour, API usage) and scores new events against that baseline. Anomalies flow into the audit ledger with explanation traces — so you can act on them, not just see them.
What it does
Learned tenant-normal
Genome trains on your historical data — no manual rule sets, no off-the-shelf thresholds.
Explanation traces
Every anomaly comes with the features that triggered it. Genome doesn't just say "weird" — it shows you the why.
Cross-domain scoring
Transactions, agent calls, login attempts, API usage — all scored in one model so cross-channel patterns surface.
Drift detection
Integrates with Tracks for ML drift; with Conscience for gate-failure spikes; with Close for ledger anomalies.
When to reach for it
- You want fraud / anomaly detection that explains its calls.
- You're shipping ML models and need drift detection that ties back to evidence.
- You operate workflows where "X agent suddenly using Y tool 100×" needs to trigger a human review.
- You want continuous anomaly scoring across finance, security, and operations in one place.
Pricing
This product is included starting from the Pro tier ($49 CAD/mo). Every higher tier includes it automatically. See the tier & product mapping for the full picture.
Read more
- Genome docs
- Conscience — gates that Genome triggers.
- Tracks — ML drift partner.