Accelerator 04 · Pharma
Regulatory & literature research assistant
Answers regulatory and scientific questions only from your document set, with a citation for every statement.
The problem
- Regulatory, QA and medical affairs teams search through SOPs, guidance documents and study reports by hand.
- General chat tools answer without sources, which is not acceptable for regulatory work.
- Knowledge sits with a few senior people and leaves when they do.
What it does
- Retrieves the relevant passages from your approved document set.
- Answers only from those passages and cites the document for every claim.
- Says plainly when the documents do not answer the question.
- Keeps an audit log of questions, sources used and answers.
Ask a question over a small sample document set and get a cited answer.
Live demo: Regulatory & literature research assistant
Synthetic data only · nothing you type is stored
Output
An answer where each statement cites a document id, plus a list of what the documents do not cover.
The demo first calls the Surge in-house AI server and falls back to a hosted open-source model if it is busy. The badge on the output shows which one answered. Limited to a few runs per visitor per day.
How it integrates
In your deployment the accelerator runs inside your environment, behind your single sign-on, and is connected to the systems below. You choose the hosting and the model.
Systems and APIs
- Document management (e.g. a validated DMS) export or connector
- SharePoint / file share crawl for approved folders
- SSO for access control by department
Data we need for a pilot
- The approved document set (SOPs, guidance, study reports)
- 50 real questions with expected answers for evaluation
Where it runs
- On-prem for confidential dossiers
- Private cloud for public-domain literature
Open-source models we start with
- Qwen3-class 14B–32B instruct model for grounded answers
- BGE-M3 embeddings + a reranker for retrieval
- pgvector or Qdrant as the vector store
Walkthrough
A short recorded walkthrough of the regulatory & literature research assistant: the input, the output and how a reviewer uses it.
Next accelerator: Catalogue enrichment + shopping assistant →Walkthrough video coming soon
We are recording a short walkthrough of this accelerator. Meanwhile, run the live demo above or book a working session.
Run the pharma accelerator on your data
In a working session we map it to your process, systems and rules, and agree what a pilot must prove before it goes live.