Draft with AI. Analyze with AI. Integrate with anything.
The EDC is open on both ends: AI accelerates study setup and analysis, and a permission-scoped API moves your data to the systems — and vendors — you already work with.
From protocol PDF to running study — and back out to analysis.
Hand the assistant a protocol PDF and it drafts the CRFs, visit schedule, and consent bindings for your data manager to review and approve — never applied blindly. Your model key stays encrypted and tenant-scoped.
Connect an AI analyst — Claude, or your own tooling — over our MCP interface: it can pull the complete, permission-scoped dataset, run the numbers, and draft the report your team finishes. The same open REST API moves data to any downstream system, and every agent call runs under your permissions, your row-level security, your audit trail. Signing and token minting stay human-only.
# authenticated by an issued token — acts as your user $ edc subjects --study 2025HZ001 --site GE-GAL A005 enrolled 88% A006 enrolled 100% e-signed A007 screening 41% 1 query $ edc query open A007 --field temp \ --text "Confirm 38.4 °C — outside range" ✓ query #1180 opened ↳ recorded in audit_log, attributed to you
Lock the database. Start the analysis the same day.
The point of an EDC is what comes out of it. Every export is tidy, coded, and standards-shaped — so your statisticians and data scientists start from clean tables, not from cleaning.
CDISC SDTM & ODM, Excel, CSV
Domain datasets for submission work, ODM XML for archival and system-to-system interchange, real Excel workbooks and CSV for everyone else.
- Explicit missing-value reasons — “not done” never reads as a gap
- Print-ready PDF casebooks per subject
IVD statistics built in
Method comparison runs inside the platform, on live data: bias, Bland–Altman limits of agreement, robust regression, PPA/NPA/OPA with confidence intervals.
- Acceptance criteria scored at entry (ISO 17593-style)
- Key risk indicators per site, continuously
Integrity travels with the data
Exports are permission-scoped and logged; signed records carry a content hash that can be re-verified independently of the application.
- Every export attributable — who, what, when
- Optional blockchain anchoring of record hashes
Name the system. We build the interface.
Already committed to a LIMS, a safety database, a statistics platform, or a sponsor portal? Tell us which. We design the interface with you — mapping, transport, and validation evidence — and ship it as part of your study setup. That's what having the development team behind the product means.
Bring your protocol and your toolchain.
We'll connect the two — and your data will arrive where it's needed, analysis-ready and fully attributable.