IN PRIVATE BETA

Machine speed. Expert control.

Augura runs the whole study, from the question to the report. The researcher makes every call and signs the result.

AI-native workspace for health data analysis

BENCHMARK

250 hours of analysis, now 2.5

We re-ran the analysis of the STOP-or-NOT published multi-center randomized trial on Augura.
The original took 250 hours over 45 days, on Augura it took 2.5 hours, with the same results.

Scope: from receipt of the analysis dataset to completed primary and secondary results.
Trial conduct, data collection and manuscript writing excluded.

Working with research teams at

“Augura is making expert-level methodological reasoning accessible to research teams.”
Person

Christopher Seymour, MD

University of Pittsburgh School of Medicine

“Augura is making expert-level methodological reasoning accessible to research teams.”
Person

Christopher Seymour, MD

University of Pittsburgh School of Medicine

“Augura compounds expert knowledge, allowing every new study to build on those that came before rather than starting from scratch each time.”
Person

Vincent Liu, MD

Chief Data Officer, The Permanente Medical Group

“Augura compounds expert knowledge, allowing every new study to build on those that came before rather than starting from scratch each time.”
Person

Vincent Liu, MD

Chief Data Officer, The Permanente Medical Group

Built for the teams
who produce the evidence

Payer dossiers, post-market studies, label expansions and publications all queue behind a small pool of biostatisticians and epidemiologists. Each study moves between four to six specialists and through SAS, R and Word before anyone can sign it. Hiring doesn’t clear the queue: the people are scarce and take years to train.

Pharma & biotech

Comparative effectiveness and safety, external control arms, subgroups and treatment-effect heterogeneity, indication expansion, HEOR, payer and HTA evidence.

Medtech, diagnostics & digital health

Clinical utility and real-world performance, post-market clinical follow-up, post-approval studies, and value evidence.

Health systems & academic medical centers

Outcomes and quality studies on your own data, ready to publish, without waiting for biostatistics time.

CROs & consultancies

Deliver more studies with the methodologists you have, with designs that stay consistent from team to team.

Fast enough for the backlog,
rigorous enough to sign

Rigor that is checked, not assumed

Formal identification, assumption audits and sensitivity analyses in every study. Automated cross-checks and expert review challenge the analysis before any result is trusted.

Auditable end to end

Every conclusion links to its source data, definitions, assumptions, analytical decisions and code. Your researchers make and sign every call.

“We review its outputs with the same scrutiny we apply to peer-reviewed research.”
Person

Michael O. Harhay, PhD, MPH

University of Pennsylvania

“We review its outputs with the same scrutiny we apply to peer-reviewed research.”
Person

Michael O. Harhay, PhD, MPH

University of Pennsylvania

Built around target trial emulation, the ICH E9(R1) estimand framework, and FDA and NICE real-world evidence guidance.

Augura does the work,
researchers make the calls

Frame the question

Turn a research question into a precise one: who, what exposure, what outcome, compared with what.

Prepare the data

Map and check every variable, and flag what’s missing or inconsistent.

Map the causes

Augura proposes the causal diagram. Your researcher edits it and locks it.

Map the causes

Augura proposes the causal diagram. Your researcher edits it and locks it.

Plan the analysis

Choose the design and the method, and test them on simulated data first.

Plan the analysis

Choose the design and the method, and test them on simulated data first.

Crash-test the assumptions

See how much it would take to change the conclusion.

Crash-test the assumptions

See how much it would take to change the conclusion.

Write the report

Tables, methods and the full decision log, ready for a reviewer, a payer or a regulator.

Built by researchers
who have run these studies for twenty years

Romain Pirracchio, MD MPH PhD
Founder, Science

UCSF Distinguished Professor of Anesthesia and Statistical AI Editor at JAMA. PhD in biostatistics; 240+ publications in causal inference.

Romain Pirracchio, MD MPH PhD
Founder, Science

UCSF Distinguished Professor of Anesthesia and Statistical AI Editor at JAMA. PhD in biostatistics; 240+ publications in causal inference.

Matthieu Legrand, MD PhD
Founder, Methodology

Tenured UCSF Professor of Medicine and world-leading critical care expert. 300+ publications in NEJM, JAMA, The Lancet.

Matthieu Legrand, MD PhD
Founder, Methodology

Tenured UCSF Professor of Medicine and world-leading critical care expert. 300+ publications in NEJM, JAMA, The Lancet.

Francois Cottin, MBA
Founder, CEO

Repeat founder who turns deep technology into markets: telecommunications, hardware, and B2B software through its acquisition by Adobe. Ex-BCG.

Francois Cottin, MBA
Founder, CEO

Repeat founder who turns deep technology into markets: telecommunications, hardware, and B2B software through its acquisition by Adobe. Ex-BCG.

Smiling man with light brown hair and stubble, wearing a khaki shirt in natural light

The private beta is open to a few more research teams

The private beta is open to a few more research teams