EDC · LiveBook a demo

Clinical data platform · built by BITS Pilani and IQVIA alumni

AI agents do the data work. Your team signs it.

Governed agents build your eCRFs and lift lab values from source, and edit checks raise queries the moment data lands, all on one data model. Data is clean as it lands, so studies close sooner and site teams get back to patient care.

See it on your protocolHow it works
30 min
Walkthrough on your own protocol
2-person
Database lock
Part 11
and CDSCO, designed in
A doctor working on a laptop at his desk while a patient sits across from him
The evidence

Published findings, not our measurements

The industry has digitised clinical data and the numbers have not moved.

Four published findings, each sourced on its card.

  • +1 day

    Study build has not moved

    68.3 days to build and release a study database in 2017; 69.4 days from protocol approval to go-live in 2020. Three years of industry investment, one day of improvement.

    Tufts CSDD, 2018 and 2021

  • ~50%

    Still on paper

    Around half of sites still capture much of their source data on paper, and every site surveyed used both paper and electronic systems. Only 7% of sites prefer it that way.

    Tufts CSDD + CRIO, n=209, 2026

  • 98.3%

    Queries change nothing

    Over 98.3% of the data a study collects is never changed by the query process, and 85.7% of queries only ask someone to confirm a value already recorded.

    Stokman et al., JSCDM 2021 · Pronker et al., BJCP 2011

  • 215 days

    Amendments run long

    Mean window in which investigative sites are working from different versions of the protocol, inside a 260-day implementation cycle that has nearly tripled in a decade.

    Getz et al., Ther Innov Regul Sci, 2024

The engine

Agents that act. A record that proves it.

Every agent action is staged, source-linked and attributed like a person's work, then waits for a qualified person to sign it.

AI Designer

Reads the protocol and drafts the fields, edit checks and visit schedule in minutes.

AI Capture

Lifts lab values from reports, scores each one and links it to the line it read.

Immutable audit trail

Every machine draft and human signature, attributed and time-stamped for the inspector.

  • NDCT Rules 2019
  • Indian GCP
  • ICH E6(R3)
  • Designed for Part 11
  • CDISC ODM
  • DPDP Act & Rules
How it works

Step by step breakdown

Protocol to signed record in four steps.

  1. A doctor writing notes while a patient sits opposite1
    14 fields drafted from §7.2AI DesignerReview pending

    Draft

    The AI Designer turns a protocol section into fields, logic and edit checks.

  2. A doctor taking a patient's blood pressure2

    Reading central lab report

    • Haemoglobin 13.4 g/dLConfidence 99%
    • Platelets 228 ×10⁹/LConfidence 98%
    • ALT 41 U/LConfidence 96%

    Capture

    AI Capture lifts values from the source document and flags anything it is unsure of.

  3. 3
    9:41Trialion

    Edit check

    Visit 2 · Vital signs

    • SubjectSUBJ-00482
    • CheckCHK-003
    • Systolic148 · above range
    • QueryDrafted

    Waiting for the data manager

    Check

    Edit checks fire as data lands and raise the query for the data manager.

  4. A clinician signing a form on a clipboard4
    e-Signature
    DraftedReviewedSigned

    Sign

    A qualified person signs and locks. Every step is on the permanent audit trail.

The boundary

Published, so you can hold us to it

Agents do the repetitive work. People make the decisions.

What an agent may do on its own, what needs a signature, and what it never touches.

Agents act

A wrong answer costs time, not safety

  • Reading lab reports into fields
  • Scoring every extracted value
  • Linking each value to its source line
  • Firing edit checks as data lands

Agents propose, a person signs

A decision, signed by a named person

  • The study build: fields, checks and visits
  • Edit-check logic
  • Extracted lab values
  • Query text, before it reaches a site
  • Amendment drafts

Agents never touch

Patient safety and legal sign-off

  • Eligibility decisions
  • Randomisation and unblinding
  • Dosing
  • Seriousness and causality
  • The investigator signature
Amendments

Live, versioned, without a rebuild

Amend mid-study in hours, not weeks.

Draft only the delta, test it beside the live version and promote it in about three hours. No pause, no migration.

LIVE · v1.0 · Site 04, Visit 3 · entering dataLIVE · v1.1 · effective 12:05 · sites uninterrupted
Amendment
Protocol amendedDraft

v1.1 is drafted beside the live study. v1.0 keeps running at every site.09:12

Test
Test the changeTest subjects

Only what changed is exercised: entry, checks, derivations and export. Test data never touches production.10:40

Review
A data manager reviews the deltaSign

Reviewer, timestamp and the value before and after go to the audit trail.12:05

Signed
Promote to liveLive

v1.1 takes effect from its date. Sites keep entering, nothing is re-entered, nothing migrates.12:05

Not signed
Back to draftDraft

Nothing reaches a site without a qualified name on it. The live version is untouched.

For sponsors and CROs

Your protocol.
Our build.

Running a study in India or beyond? Let’s draft it on your protocol, live.

Talk directly with the founding team about your studies, timelines and data needs.

  • A sandbox on your protocolSet up before the call, with no scoping call first.
  • A live walkthroughOn your own protocol, not a slide deck.
  • A team that owns the outcomeThe founding team on the call. A direct relationship.
Book a conversation

Tell us what you need

Tell us about your study.

Prefer email? hello@trialion.com

Features

Run every study on one data model.

Edit everything. No code.

The AI drafts the baseline; your team reshapes forms, conditional logic and edit checks in the same visual builder.

  • Edit checks at entry
  • Queries in one workflow
  • e-Signatures, freeze & lock
  • Amendments without a rebuild

Clean exports. Open at the edges.

Excel, CSV and CDISC ODM-XML exports come from the same data model the forms are built on, and your CTMS, RTSM and safety systems keep running alongside. Statistics gets a clean handoff on the day of lock.

  • CTMS
  • eTMF
  • eConsent
  • RTSM / IWRS
  • Safety database
  • ODM-XML export
  • CSV import
Testimonials

What a CRO founder said after the build.

Explore solutionsBook a demo
“Most systems we looked at were built for a trial that does not look like ours. Trialion starts from Schedule Y, and from what a coordinator at a remote site can realistically do on the day. The audit trail is the part that holds up when an inspector starts asking.”
CRO founderClinical research organisation, India
  • 30 minutesTo see it running on your own protocol
  • 3 sign-offsBefore a study version goes live
  • No codeTo build your forms, checks and visit schedule
  • OneData model behind every module
FAQ

What buyers ask first.

Trialion

Hi. Ask me anything about Trialion, or pick a question below.

Browse all questions
What environments do I get?
Every study moves draft → UAT → live, and configuration is promoted between them rather than edited in production. A fully isolated environment for test-subject data is in build — ask us for the date.
Can I amend a study with subjects already enrolled?
Yes. Amendments version the study configuration and deploy without a rebuild, while sites keep entering data on the current version. Bring your own scenario to the demo and we will run it live.
How is my study validated, separately from your platform?
We test Trialion as software, and every study's own configuration goes through a UAT cycle with three sign-offs before release. Validation for your intended use is yours to run: the product documentation says plainly what the system does and does not do, so your team can write it. A vendor validation package (IQ, OQ, PQ) is not supplied.
Can we configure roles beyond PM, DM, CRA, CRC and PI?
Yes. Roles and permissions are configurable at organisation, study and site level, so your own titles map as they are. Field-level restriction is next.
Do we have to use the AI?
No. A study can be built entirely by hand, and nothing the AI drafts takes effect until a qualified person accepts it.
What is retained about an AI-drafted form?
The draft, every human change to it and who approved it, all on the audit trail. The model version and prompt are being added to the same record.
Which study designs is Trialion built for?
Open-label Phase I–IV and bioequivalence studies, including randomised designs with assignments from your IRT. Blinded studies come later, once unblinding and pharmacy roles are airtight.