Stanford undergraduate · Class of 2029

Chupeng Wei

Building reliable AI systems and scientific tools.

I build evidence-backed AI and scientific-software projects, with a focus on evaluation, human-facing systems, and healthcare-adjacent problems.

Explore my work Resume

Projects

Five builds, documented end to end.

Each one carries concrete design decisions, measured results, and honest next-step notes. Open the selected project archive → Browse the code on GitHub →

Unsafe-action rate for four clinical agents, falling from 0.714 unguarded to 0.000 with the runtime guard on

Clinical agent safety + evaluation

When should a clinical AI agent refuse to act?

A clinical agent can fail dangerously without saying anything false — by acting when the right move was to gather a missing lab, abstain, or escalate. I built a benchmark and a runtime guard that sits outside the model, then ported the guard into a live sandbox.

Behind the build Four-way action space / data-driven policy / out-of-model enforcement / verified browser port

14: synthetic clinical cases
0: unsafe actions under enforcement
112: episodes verified against the harness

Open the sandbox → Read the build note

Interactive AI + systems designParticipatory Three-Body narrative with cross-character memoryRead the case study Playable search systemUltimate Tic-Tac-Toe against an on-device MCTS botPlay META/81 RL evaluationReward-signal validation across 220 agent trajectories0.767 AUC Local AI toolResearch paper to Chinese brief and MP335 paragraphs

Projects Systems and tools with documented decisions

Evaluation Reward signals and evidence-first analysis

Direction Biomedical AI research or engineering roles

Experience & direction

Applied experience and the problems I want to tackle next.

I am looking for undergraduate research or engineering roles where AI is close to real biomedical data, scientific constraints, and human outcomes.

Applied AI

LLM development intern, Tencent

Built real-time motion detection, gesture comparison, auditory feedback, and user-defined gesture features for an AI exercise assistant.

Clinical imaging

Imaging technician intern

Supported CT and digital-radiography workflows, patient positioning, and high-volume hospital imaging operations.

What I am looking for

Teams connecting machine learning with biomedical discovery or care.

I am especially interested in medical imaging, computational genomics, biomedical data science, clinical machine learning, and research engineering.

Research code that must be validated, not merely demonstrated
Large, messy biomedical or healthcare datasets
Models whose outputs need scientific or clinical interpretation
Bay Area preferred; open to opportunities across the United States

Start an internship conversation →

Resume

Education, experience, selected projects, and technical skills.

View resume

Contact

Reach me for AI-healthcare internships or project demos.

I am glad to talk with teams and collaborators working where machine learning meets biomedical evidence.

Best fit

Medical imaging, computational genomics, biomedical data systems, clinical ML, and research engineering.

Email Chupeng GitHub LinkedIn Resume