Skip to content

In development

AI for clinical-trial quality control.

We’re building Akratos to help biostatistics teams review statistical programming and investigate discrepancies.

How it works Meet the team
Workflow conceptFor biostatistics teams
  1. 01

    Study inputs

    Analysis plan, specifications and data

  2. 02

    Independent check

    A separate QC implementation

  3. 03

    Findings for review

    Discrepancies to investigate

A proposed workflow, from the study’s requirements to findings the team can investigate.

An independent second look.

Our initial focus is an AI second programmer for clinical-trial statistical programming. The aim is to give biostatistics teams another way to check their work, with people reviewing the findings.

The starting point is the study itself: its analysis plan, specifications and data. A separate QC implementation would be compared with the study outputs.

A separate check

Built around an independent review of statistical programming.

Clearer findings

Focused on helping teams investigate differences.

Human review

Designed to support the people responsible for the work.

Portrait of Nicholas O’Connell, co-founder of AkratosPortrait of Julius Reich, co-founder of Akratos

Nicholas O’Connell & Julius Reich

Meet the people building Akratos.

We both study at Bocconi University in Milan. Our backgrounds span entrepreneurship, applied AI and business research.

Our team

Selected for TEF Ignition, Autumn 2026.