About Reprosia

Why we built
this platform.

Pharmaceutical data science has a reproducibility problem. Reprosia exists to fix the infrastructure layer that every regulated analysis depends on — and nobody else was building.

Our mission
"Make reproducibility and regulatory compliance the path of least resistance for every pharma R&D team."

Reprosia started with a simple observation: most clinical analyses cannot be reproduced. Not because the science was wrong, but because the R environment wasn't documented, the package versions weren't locked, and nobody wrote down exactly what happened between raw data and submission output.

This is a solvable problem. The tools exist in open-source R. The regulatory requirements are well-understood. What was missing was a platform that connected them — making compliance automatic rather than aspirational.

We built Reprosia to be that platform. AI-native, built on open-source foundations, designed from day one for the teams that cannot afford errors: pharma, biotech, and CRO statisticians working under FDA and EMA oversight.

We are not a regulatory consultancy that happens to use software. We are a technology company that deeply understands regulation. The distinction matters: our goal is to encode regulatory expertise into software that scales — not to bill consulting hours indefinitely.

The open-source foundation — reproducr, regulog, lineager, and estimandr — is our commitment to the community. These packages solve real problems for individual statisticians and teams. Reprosia is the platform that makes those solutions work at enterprise scale, with the AI audit layer, environment management, and submission tooling that regulated environments require.

We are based in Brussels, Belgium, registered as ReproStats BV. We work with pharma and biotech teams across Europe and globally. EU data residency is built in, not bolted on.

This is early days. We are working with a small number of teams in private early access. If reproducibility is a problem you are trying to solve, we want to talk to you.

0.2.1
reproducr current version on CRAN
4
open-source packages in the ecosystem
97%
test coverage across all packages
EU
data residency — Belgium, GDPR compliant
What we believe

The principles that
guide every decision.

01

Open source first

The reproducibility problem belongs to the community. Our packages are open-source, peer-reviewed, and will remain so. The platform adds enterprise capabilities — it does not hide the science behind a paywall.

02

Compliance is not optional

We don't build workarounds for regulatory requirements. We build tools that make compliance easier to achieve — and harder to accidentally skip. The goal is a world where doing the right thing is also the easiest thing.

03

Statisticians are scientists, not sysadmins

Clinical statisticians should be thinking about their estimand strategy, not debugging R package dependencies. Reprosia handles the infrastructure so people can focus on the science.

04

Trust is earned by transparency

In regulated environments, every claim must be verifiable. We hold ourselves to the same standard: our validation methods are documented, our audit trails are complete, and our outputs can always be traced to source.

05

AI in service of rigor

We use AI to make compliance more thorough, not less rigorous. AI is the audit engine — it catches what humans miss, at a scale no manual process can match. It does not replace regulatory judgment; it extends it.

The team

Built by someone who
lived the problem.

NP
Ndoh Penn
Founder · ReproStats BV & Reprosia

Biostatistician and R developer based in Brussels, Belgium. Before building Reprosia, Ndoh spent years working on clinical trial analyses — watching submissions fail not because the science was flawed, but because the computational infrastructure around it wasn't documented or reproducible.

The experience led to a clear conclusion: the tools to fix this existed in open-source R, but nobody had assembled them into a platform that regulated teams could actually use. ReproStats started as that toolkit — four open-source packages covering audit, provenance, logging, and estimand specification. Reprosia is the platform layer that makes them enterprise-grade.

Ndoh holds deep expertise in ICH E9(R1) estimand methodology, 21 CFR Part 11 compliance for R environments, and CDISC data standards for regulatory submissions.

Open-source foundation

ReproStats: the foundation
beneath Reprosia.

Reprosia is built on four open-source R packages, peer-reviewed and available on CRAN and GitHub. The platform adds the AI audit engine, environment management, and enterprise workflow — on top of a foundation that belongs to everyone.

reproducr
On CRAN
Script audit, environment certification, risk scoring and drift detection for regulated R analyses. The core reproducibility engine.
regulog
In development
21 CFR Part 11-compliant electronic audit logging for R. Tamper-evident, user-attributed, timestamp-verified action records.
lineager
In development
Row-level data provenance for clinical datasets. Track every transformation from raw SDTM to analysis-ready ADaM with full traceability.
estimandr
In development
ICH E9(R1) estimand framework for R. Define, document, and implement your estimand strategy directly in your analysis pipeline.
Explore reprostats.org →
Get in touch

We are building this
with you.

Early access is limited. We work directly with every team to understand their environment and tailor the platform to their regulatory context.

Or email ndoh@reprosia.com