Professor Review
A rubric-aligned summary of the AI Medical Assistant, with links to the page behind each result.
Download final submission
Enter the access code to download the final submitted presentation.
RAG groundedness
5.0 / 5
base LLM 3.6
RAG relevance
4.8 / 5
final pipeline; comparison run 4.8 / 5.0
Chunks indexed
11,741
4,114 pages
Models
gpt-4o-mini
judge gpt-4o
Question answering using LLM
Base model answers to all five questions with observations.
Open evidence
LLM with prompt engineering
Clinical system prompt, parameters and answers for all five questions.
Open evidence
Data preparation for RAG
Loading, cleaning, chunking, embedding, Chroma vector database and retriever.
Open evidence
Question answering using RAG
Grounded answers with page citations, plus five fine-tuning runs and a six-configuration grid.
Open evidence
Output evaluation
Separate judge model scoring groundedness and relevance, plus a clinical checklist.
Open evidence
Actionable insights & recommendations
Rollout guidance, safeguards and the 2011-source limitation.
Open evidence
Evaluator review workflow
Enter the review access code to capture evaluator notes and store them as a numbered version alongside this submission.