Cited feasibility for CROs & sponsors

Cited feasibility your CRO can bid with — and defend.

Protocol in, cited go/no-go verdict out — in hours. Every number traces to its public source, on a Part 11 audit trail. White-label and ready for your bid.

Book a demoSign in
White-label & dedicated-cluster readyVolume API for bid pipelinesNever trains a model on your protocol
Sample verdictillustrative — not real data

ENDO-2b · Endometriosis Phase 2b

Feasibility verdict · every number cited

CONDITIONAL GO
PTRS
58%Open TargetsTarget–disease association evidence integrating genetics, pathways, and known drugs — feeds PTRS scoring.Click to inspect & verify ▸
Eligible cohort (US)
24,300AACTRelational, queryable database of all ClinicalTrials.gov content — the analytical backbone for site and enrollment statistics.Click to inspect & verify ▸
Median enrollment
14.2 moClinicalTrials.govRegistry of public and private clinical studies worldwide — enrollment, status, sponsors, sites, and outcomes.Click to inspect & verify ▸
Est. per-site budget
$1.18MPubMedBiomedical literature index of 36M+ citations — used to ground indication baselines and standard-of-care evidence.Click to inspect & verify ▸
Part 11 tamper-evident · hash e7a1…9f

Built on public data sources

ClinicalTrials.govPubMedFDAEMAOpen TargetsAACTChEMBL

Public, re-verifiable sources — click any number to check it yourself. Your protocol is never used to train a model.

These 7 backbone registries are part of ~50 live connectors feeding the engine.

The moat

The opposite of a black box.

One verdict, read two ways: defensible enough for your board, transparent enough for your engineers.

For the buyer

  • Defensible to your board and the FDA
  • Hours, not a six-week cycle
  • Built on public data anyone can re-verify
  • No proprietary lock-in

For the builder

  • Every number resolves to a public-source URL
  • Deterministic scorers, not model guesses
  • Part 11 tamper-evident hash chain
  • Private EU-region inference — your protocol is never used to train a third-party model

The problem

A multi-million-dollar go/no-go shouldn't rest on a report you can't audit.

If you run a CRO bid team, you know the math: a $100k, six-week feasibility cycle is fine when you win the bid and unsustainable when you lose it — and a sponsor can still pick the number apart. The same is true for a biotech or academic IST unit making an irreversible bet. Today that evidence arrives slow, opaque, and impossible to check.

Speed

A six-week CRO feasibility questionnaire is too slow for a go/no-go decision that keeps moving. By the time the report lands, the window has narrowed.

Defensibility

When your board — or the FDA — asks where a number came from, "the model said so" is not an answer. You need a verdict anyone can trace back to a public source.

Cost

Bespoke feasibility consulting costs five or six figures per protocol — before you have even decided whether the trial is worth running.

How it works

Protocol in. Cited verdict out. In hours.

1

Ingest

Upload your protocol PDF. An LLM in our private EU-region inference environment extracts the endpoints, eligibility criteria, and trial design — your document is never used to train a third-party model.

2

Pull public data + score

We query ClinicalTrials.gov, PubMed, FDA, EMA, Open Targets, AACT, and ChEMBL, then run deterministic scorers for PTRS, budget, site density, and cohort reachability.

3

Cited verdict

You get a verdict where every number carries its source plus a tamper-evident Part 11 hash. Export the whole thing for your board or the FDA.

How it compares

Faster than a CRO. Defensible unlike a chatbot.

A feasibility verdict that's quick to get and impossible to wave away.

CapabilityCRO questionnaireBlack-box AIClinical Trial OS
Turnaround4–6 weeksMinutesHours
Every number traced to a public source
Deterministic scorers (re-runnable)
21 CFR Part 11 tamper-evident audit chain
Your protocol never trains a model
Cost per protocol5–6 figuresLowFlat per-run

For developers

A feasibility verdict from one API call.

POST a protocol, get back a cited, scored, Part-11-audited verdict.

  • One POST, a fully cited, scored verdict back.
  • Token auth + per-tenant isolation — your data never trains a model.
  • Part 11 audit hash on every response.
Full API reference →
POST /v1/feasibility
curl https://$FEASIBILITY_API_HOST/v1/feasibility \
  -H "Authorization: Bearer $TOKEN" \
  -F [email protected]

# response — every figure carries its source
# verdict:        CONDITIONAL GO
# ptrs:           58%   (source: opentargets)
# eligible_us:    24,300 (source: aact)
# audit_hash:     e7a1…9f  (part-11 chain)

Private beta — the public API is not yet publicly available.

For your agents

MCP · feasibility.score
> tool  feasibility.score
  protocol = "ENDO-2b Phase 2b endometriosis"

# result — every figure carries its source
# verdict:     CONDITIONAL GO
# ptrs:        58%   (source: opentargets)
# audit_hash:  e7a1…9f  (part-11 chain)

Private beta — connect from Claude, Cursor, or your own agent.

Use it from Claude, Cursor, or your own agent.

Connect our MCP server and ask for a cited feasibility verdict in plain language. Citations and audit hash included.

Connect the MCP server →

Founding cohort

Onboarding a small founding cohort of design partners.

We are taking on a handful of CRO and biotech teams to run real feasibility work with us — before we scale. If that is you, the door is open.

Run your real protocols

Bring live feasibility questions, not toy examples — every verdict is cited back to public data you can check.

Shape the roadmap

Tell us where a verdict is wrong. Founding-cohort feedback sets what we build next and earns preferential pricing.

Named only with consent

We will never show your logo or name until you have explicitly agreed to be a public reference.

Apply for founding-cohort access

Pricing

Per-protocol or annual. No per-seat.

Pay per feasibility run, or go annual for high-volume (fair-use) protocols across your team.

See pricing →

FAQ

Questions, answered honestly.

You give it a protocol PDF (or synopsis). It returns a feasibility verdict — PTRS, site density, budget envelope, and cohort reachability — where every single number is cited to its public source (ClinicalTrials.gov, PubMed, FDA, EMA, Open Targets, AACT, ChEMBL). Nothing is an unsourced "AI guess".

See a verdict you can actually check.

Send us a protocol. We'll return a fully cited feasibility verdict you can trace, line by line, back to public data — yours to white-label, defend in a bid, or hand to a regulator.

Book a demoSign in