Helix — 1‑page investor memo (decision trust infrastructure)

Canonical one‑liner

Helix is the system of record for in‑silico genome‑editing design decisions: deterministic runs, policy‑gated claims, and offline‑verifiable evidence bundles that explain drift instead of hiding it.

Verdict (blunt)

This is a real $1B software thesis if and only if Helix stays positioned as decision trust infrastructure (verifiability under adversarial scrutiny), not a “genome editing platform”.

The problem (what incumbents don’t solve)

Most orgs can record activity (notebooks, ELN/LIMS, CI logs). They cannot produce a verifiable record of decisions:

• Which policy was applied (pinned + reviewable)?
• Which semantics/scoring version was used (semantic hash)?
• Which environment executed it (backend fingerprint, determinism class)?
• Can a third party replay and detect drift offline (trust check + verifier)?

The solution (what Helix makes possible)

Helix turns “decision trust” into a deterministic artifact contract:

• Scientific contract + trust kit: determinism classes, taint/trust labels, pinned policies, and a verifier that fails closed.
• Receipts as products: signed verdicts + reference validation bundles as reusable trust anchors.
• Divergence as a feature: run graphs + minimal causal subgraphs for “why did this change?” plus a shareable proof bundle.

What we sell (the revenue spine)

Teams/Registry is a programmable acceptance boundary. It distributes pinned policies and enforces the same gates as public reference bundles, inside an enterprise environment.

Not collaboration software; not an ELN/LIMS replacement.

Why this is hard to copy (the moat)

Helix already combines:

1. Contract + verifier that binds policy + semantics + environment + outputs
2. Governance gating (fail‑closed export, side‑channel control)
3. Signed bundles/receipts that downstream systems can treat as acceptance gates
4. Divergence tooling that makes drift diagnosable (reviewable proof bundles)
5. Formal-ish DAG correctness checks (VeriBiota)

Near‑term plan (milestone shaped)

• Now: make CI say no — “CI will not accept this result unless Helix verifies it.”
• 0–3 months: make adoption the default — copy‑paste CI templates + docs, expand signed validation packs, enforce contract identity stamping everywhere.
• 3–9 months: ship drift UX — one‑screen CLI summary, one‑pane Studio compare flow, and a verifier‑clean divergence proof bundle.

Success metrics (measured in CI artifacts)

• Median “time to explain drift” (two bundles → causal frontier) under 2 minutes on a cold machine.
• % of bundles that are pinned‑policy compliant with a complete backend fingerprint for the declared determinism class.
• Zero “silent drift” regressions: changes that affect declared checks fail a gate or produce an explicit, reviewable diff bundle.

Non‑negotiables (protect the thesis)

• Fail closed always. Never add a convenience path that bypasses trust.
• Keep offline verification sacred. Verification must not require Helix servers to be trusted.
• Don’t oversell biology. Sell determinism, replay, and verification; genome‑editing simulation is the first proof.