CV

ML Research Contributor

MedARC

• Drove 37% of the gain over baseline on MedARC’s NanoPath benchmark; built an LLM autoresearch pipeline with a dual-branch anti-overfitting architecture.
• Built an independent local pathology-model robustness pipeline similar to WAIV’s work, surpassing WAIV-reported results on 7 of 8 tasks and improving scanner-noise robustness by 34%.
• Post-trained Qwen3.6-27B with GRPO, cutting reasoning-token usage by 30% without benchmark loss; served locally with vLLM.

Simons Summer Research Fellow (Computational Genomics)

Yurovsky Lab, Stony Brook University (Simons Foundation)

• Architected a DNABERT-based prognostic model using Cox regression on non-coding regulatory DNA mutations from 185 patients to predict glioblastoma survival.
• Identified significant age-related survival signals and racial disparities in mutational scores.

Pathology AI Research Intern

AI in Medicine & Computational Biology Lab, Weill Cornell Medicine

• Conducted foundation-model evaluation across TCGA and internal cohorts by benchmarking 12 pathology models (UNIv2, Prov-GigaPath).
• Found a 2x out-of-distribution error increase (MAE ~0.10 to ~0.20) in tumor-purity prediction, highlighting clinical deployment risk.

Lead Technical Developer

MathLinks.org

• Led engineering for a high-concurrency competition platform serving 150,000+ users across 11 partner organizations.
• Built React/Node.js authentication, submission, and leaderboard workflows.
• Secured $20,000 in seed funding (Hudson River Trading sponsorship, Innovate901 1st Place).

Computational Oncology Research Intern

Advanced Computing & Oncology Lab, Memorial Sloan Kettering

• Built a CPU-only local LLM pipeline with LLaMA 3.1-8B, Qwen3-4B, and DSPy/MIPROv2 to classify 370 clinically annotated lung cancer CT reports; coauthored a referral-event analysis abstract accepted at the ACRO 2026 Summit.
• Applied Cox regression and survival analysis to 230 patients’ clinical and demographic records to identify correlations between percent thymic tissue and NSCLC outcomes.

Full-Stack Developer

Cyberlinc, Inc.

• Engineered a secure full-stack crowdfunding platform using Flask and SQLAlchemy with integrated payment gateways for project funding.

B.S.

University of Illinois Urbana-Champaign

Computer Science

• Expected May 2029

William Fremd High School

• AIME Qualifier (4x), Simons Fellow
• Relevant Coursework: Linear Algebra, Multivariable Calculus, AP Statistics