Document Processing
Novo Nordisk needed to reduce the 12-week, multi-team effort required to produce compliant clinical study reports without sacrificing quality or regulatory rigor.
Reported impact
NovoScribe reduced CSR creation from 12 weeks to 10 minutes, improved output quality, reduced staffing requirements, and accelerated time-to-market for new treatments.
Novo Nordisk created the NovoScribe platform to automate Clinical Study Report production with a retrieval-augmented generation architecture built on MongoDB Atlas Vector Search, Amazon Bedrock, and LangChain. Statistical outputs from clinical trials, report templates, and supporting documentation are prepared as structured tables, figures, and vectorized content inside MongoDB Atlas. LangChain orchestrates multiple Bedrock-hosted foundation models, including Claude 3 and Titan, as well as the company's private ChatGPT instance, so teams can switch models without rewriting application code. Atlas Vector Search supplies the relevant template sections and report evidence back into the prompting flow, grounding generation in the right trial context. The system therefore replaces slow manual compilation, merging, and QA-heavy handoffs with a governed AI pipeline that drafts complete CSRs in minutes, while still allowing Novo Nordisk to compare model behavior and iterate quickly as requirements evolve.
Clinical trial data is analyzed into tables and figures, templates and evidence are embedded in MongoDB Atlas, LangChain orchestrates Bedrock-hosted models such as Claude 3 and Titan, and MongoDB Atlas Vector Search feeds the right context back into the prompts so NovoScribe can assemble full clinical study reports in minutes.
NovoScribe uses the Claude 3 and Titan foundation models hosted by Amazon Bedrock, alongside the company’s own private instance of ChatGPT.
“A CSR usually takes around 12 weeks to compile, involving a multidisciplinary team of statisticians, scientists, and technical authors.”
“We’ve reduced the time taken to create Clinical Study Reports from 12 weeks to 10 minutes, with higher quality outputs and a fraction of the team.”
“With NovoScribe, we are the first in the industry to generate a complete Clinical Study Report in minutes rather than weeks.”
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
MongoDB Sample May 23, 2025 Novo Nordisk Novo Nordisk needed to shorten the slow and error-prone process for producing clinical study reports while keeping document generation deterministic and regulator-friendly in a tightly controlled pharmaceutical… "MongoDB has helped us cut writing times by 90%" MongoDB Atlas 
Google Cloud Sample Archived · Oct 9, 2026 Angelus Founded by a dentist whose first product was based on a patent, Angelus competes on differentiation, developing bioceramic products and fiberglass post lines that require clinical proof, intellectual property… 50,000 pages of proprietary data now powering AI-generated documents Gemini MongoDB Sample Oct 8, 2026 Navisio AI Rapid adoption of Navisio AI's platform triggered severe latency spikes, elevated CPU and RAM consumption, and costly vertical scaling. As it prepares to scale to 1,000 by the end of the year, and eventually toward 5,000 users, the team is… MongoDB Atlas · MongoDB Atlas Vector Search 