For AI-driven trading desks, HFT, and quantitative funds
Trading agents need deterministic replay of every executed trade, sub-millisecond decision latency, and an audit trail that survives regulator review. The cost of a memory-poisoning incident in production is measured in basis points, not engineer-hours.
Edge mode and determinism ship today — both are production grade. Memory Time Machine for execution audit and compliance officer hooks for high-stakes trade interception are in Beta. Quant teams joining the design partner program get NDA-channel access.
Important — compliance evidence: /v1/check verdicts are not included in the audit trail and do not generate W3C Verifiable Credentials. For compliance-grade evidence (HIPAA, GDPR, EU AI Act, FDA 510(k), NIST AI RMF), use /v1/preflight, which produces full audit-log entries and signed W3C VCs per verdict. The /v1/check endpoint is suitable for experimentation and high-frequency agent gating only.
Production-grade, in active use, ready for your stack today.
Zero-dependency in-process scoring. Runs in the same process as the trading agent. Sub-millisecond p99 latency. No network call between the agent and the safety check. Air-gappable for isolated dealing rooms.
pg_override_disabled + cfg_checksum mode produces bit-perfect replay of any historical decision. Regulator audit, internal compliance review, post-incident root cause — all reconstructable.
Jøsang opinion algebra for combining confidence signals from multiple data sources (market data, internal models, risk overlays). Byzantine-fault-tolerant aggregation when sources disagree.
Cryptographic attestation of memory provenance per tenant. The chain-of-custody for any decision is signed by your tenant key — never the vendor's.
Returns the projected day a given memory state will cross the BLOCK threshold. Trading desks pre-schedule strategy refresh ahead of memory degradation — predictive, not reactive.
In active development with design partner input. Not yet GA — but real, not vaporware.
Deterministic replay of the full decision chain leading to a trade, with counterfactual branching. Useful for post-mortem on flagged trades, regulator queries, and internal model retrospectives.
Status: core replay works against test workloads. The compliance-export format is being shaped with design partners (regulator audit standard varies per jurisdiction).
Pause an agent's execution mid-decision for human review on flagged high-stakes trades. Compliance officer hooks into the SSE decision stream and can veto before commit. Standard SSE protocol; the trading-desk-specific framing is the Beta part.
Status: SSE streaming is GA; the human-veto callback handshake is in test. Design partners get the first integration.
Early Access is not a feature flag — it is a structured program. Design partners get founder-direct access, priority on roadmap input, and locked-in pricing through the Beta period.
Slack channel with the founder. No ticket queue, no AE filter. You hit a wall, you hear back same day.
Beta features ship based on what design partners actually use. The next Beta milestone is decided by the first 5–10 customers, not a marketing committee.
Your Beta-period pricing is locked through general availability. If the GA tier list moves up, you stay on the partner rate for 24 months.
In return: candid feedback when something does not work, a written case study at GA (with your approval and disclosure terms), and a willingness to live with Beta-grade rough edges.
No NDA required to apply. We respond within 48 hours.
Pricing tier for funds that cannot publicly disclose vendor relationships. Long-horizon commercial work; flagged here for honesty, not committed roadmap.
More on the long-term direction →Tier badges across the page surface this distinction. No category was glossed over.
If you are a quant and trading teams and want to shape what comes next, write us. If you would rather wait for GA, the page is here when you are ready.