Aspiring AI Scientist

Shuzhen Zhang

M.S. Statistics student at the University of Washington, building practical ML systems with statistical discipline.

View projectsAbout me
Portrait of Shuzhen Zhang

Learning Path

Learning PathUniversity of Illinois Urbana-ChampaignB.S. Mathematics & StatisticsIllinois Geometry Lab2024 IML Research AwardUniversity of WashingtonM.S. StatisticsManifold LearningProf. MarinaMetricDistortion paperKeyword AISummer InternATLAS Internship Program2024 ATLAS Leadership ScholarK2DataSummer InternAmazonJr. Applied Scientist Intern

University of Illinois Urbana-Champaign

B.S. Mathematics & Statistics

Illinois Geometry Lab

2024 IML Research Award

Keyword AI

Summer Intern

ATLAS Internship Program

2024 ATLAS Leadership Scholar

K2Data

Summer Intern

Manifold Learning

Prof. Marina

MetricDistortion paper

University of Washington

M.S. Statistics

Amazon

Jr. Applied Scientist Intern

Selected Work

Featured project

Agent simulation trace showing model behavior across long context

AI Simulation with Mistral + Mixtral (ATLAS Showcase)

Prompt and task-design experiments with Mistral 7B and Mixtral 8x7B, reducing invalid action outputs on constrained assignments.

LLMMistral 7BMixtral 8x7BPrompt EngineeringSimulation
Case StudyGitHub

Research

Making embeddings easier to trust

Publication

Interactive Visualization of Metric Distortion in Nonlinear Data Embeddings

Briefings in Bioinformatics, 2026

Read paper

Research Questions

  • Multi-source hallucination

    How can we reduce hallucination when a model must reconcile multiple input sources?

  • Long-task memory

    How should long-short memory systems be designed when tasks are long but context windows are limited?

  • Embedding reliability

    How can we quantify embedding reliability across random seeds and data shifts?