An independent publicationEvery case sourced, dated & graded A–DGet the weekly Casefile →
Grade BproductionSource date Oct 17, 2026· Last verified Oct 17, 2026

How is AstraZeneca's Evinova using an AI agent to draft regulatory software documentation?

Evinova (AstraZeneca group) · Pharma, Biotech & Life Sciences

Header image: A developer coding on a laptop in a dark room

Evinova, a health tech company in the AstraZeneca group, uses Devin to draft regulatory documents for its clinical-trial software modules, work that took 35 to 40 hours of cross-functional time. Cognition reports structured first drafts at roughly 90% accuracy, with senior engineers shifting to reviewers.

Reported outcomes

first-draft accuracy
roughly 90%
Source [1] · self-reported
prior effort per module
35 to 40 hours of cross-functional coordination
Source [1] · self-reported

Outcomes are reported exactly as stated by the source; we have not independently audited them.

How it works

  1. Devin assembles context from code and systems across teams.
  2. Produces structured first drafts of regulatory documents.
  3. Senior engineers review and approve.

Risks & compliance

  • Regulated (GxP) software documentation requires qualified human review
  • Vendor-published results

Healthcare cases cover operational and commercial workflows. Nothing here is medical advice or a claim about a device's or drug's efficacy.

Build this yourself

Tools mentioned & how to build it

Disclosure: Some links may earn us a commission; it never affects grades. Tools are listed by fit to this case, never by payment.

Tools mentioned

  • Devin (Cognition)
    Autonomous AI software engineer from Cognition.

Build a version yourself

  • Notion
    Docs and wiki with Notion AI; a home for your agent playbooks and SOPs.
  • n8n
    Open-source workflow automation with AI agent nodes; self-host or cloud.
  • Make
    Visual no-code automation with AI agents; good for intake, routing and follow-up flows.

All tools by category →

Sources

[1]Cognition (Devin)· vendor customer story· Undated (accessed Oct 17, 2026)
“Devin produced structured first drafts at roughly 90% accuracy, shifting senior engineers from primary authors to reviewers”
Evinova and Devin

Grade rationale: Named organization with a quantified outcome from the vendor or the organization itself (self-reported). Automated check on Oct 17, 2026: 1/1 quotes and 2/2 outcome claims matched the source. Compiled by the Agent Casefile research desk (AI-assisted; see methodology). Spot an error? Request a correction.

Keep reading

Related cases