EvolvingScholar
EvolvingScholar is an AI "research trainee" that learns the way a real medical trainee does — mentor-guided and improving across a *sequence* of gene–disease projects.
Core trick
Stores the Scholar's concept model, question logic, and method repertoire as versioned external files that update between projects, not as prompt growth.
Description
EvolvingScholar is an AI "research trainee" that learns the way a real medical trainee does — mentor-guided and improving across a *sequence* of gene–disease projects. Given a gene and a disease, the Scholar generates research questions, reviews the literature, analyzes synthetic EMR data to test gene–disease associations, and proposes testable hypotheses. At checkpoints it "meets its PI" and receives structured, rubric-based feedback that drives its growth before the next project. The ultimate goal is to evaluate and compare AI Scholar Agent evolving trajectory growth, rather than autonomous co-scientist.
The system is built as three nested loops: **Loop A** does the research for one gene–disease pair; **Loop B** injects structured PI-mentor feedback; **Loop C** — the actual research contribution — rewrites the Scholar between projects. The core design bet is that **growth is not prompt accumulation**: cross-project learning happens only through updates to external, typed, versioned artifacts (the concept model, question-design logic, method repertoire, and earned EPAs), never through an ever-swelling prompt. This means "the Scholar evolved" is a recorded git diff, competency is read out on the clinical Entrustable Professional Activity (EPA) entrustment ladder (observe → independent → supervise), and the evolution measurement is free of the context-length confound that grow-the-prompt systems suffer.
**What we found (working pilot, v0.1.0):** two scholars — one on the Claude Agent SDK with rich tools, one on the raw Messages API with minimal tools — each completed a full A→B→C run on the first project (TTR / hereditary transthyretin amyloidosis). The API scholar reached entrustment level 2 with 4 earned EPAs on run 1. Both share one review-and-evolve harness but keep separate, endowment-coupled experience stores with no cross-pollination.
**Why it matters:** the AI Scholar is a controllable model system for a developmental process — how research questioning and disease conceptualization evolve — that is knowledge-entangled and nearly impossible to isolate in humans. Longer term it's a first step toward a self-taught AI clinical expert, the EPA based AI application in medicine, and a testbed for how AI can serve in medical education and research.
PS: slides https://dlmp-niu.github.io/EvolvingScholar/demo/talk/index.public.html
README excerpt
# EvolvingScholar **Guided self-evolution of an AI Scholar Agent through sequential gene–disease projects.** > ⚠️ Working title / directory name is provisional and may be renamed. Git history is name-independent, so a rename is safe — see [`docs/project-structure.md`](docs/project-structure.md). ## What this is An **AI Scholar Agent** (the **Scholar**) that learns the way a real trainee does — mentor-guided, improving across a *sequence* of gene–disease projects — rather than acting as an autonomous co-scientist. Given a gene and a disease it generates research questions, reviews the literature, analyzes (synthetic) EMR data to test associations, and proposes testable hypotheses. At checkpoints it "meets its PI" and receives structured feedback that drives its growth before the next project. **The question this project actually asks:** *how does research questioning and disease conceptualization evolve* — studied across a human reference panel (college intern → master's → medical student → resident) and an AI Scholar across cycles. The AI is a **controllable model system** for a developmental process that is knowledge-entangled and hard to isolate in humans. ## The core mechanism — three nested loops - **Loop A — Research activity** (inner): the Scholar does the research for one gene–disease pair. - **Loop B — Scholar ↔ PI** (middle): structured mentor feedback at checkpoints. - **Loop C — System update** (outer): rewrites the Scholar's **external, typed, versioned art