About Me

CS student and ML engineer at UW — working on sparse autoencoders, SSL representation theory, and mechanistic interpretability

S² Representation Geometry · alignment vs uniformity · drag to rotate

Background

I'm a Computer Science student at the University of Washington's Paul G. Allen School of Computer Science & Engineering, maintaining a 4.0 GPA while doing research in mechanistic interpretability and representation theory.

Currently I'm working as an ML Research Engineer at a stealth AI startup, investigating how CNN inductive biases shape SSL representation geometry — diagnosing alignment, uniformity, and dimensional collapse across architectures.

I was previously a Research Fellow at Algoverse AI, where I published RT-TopKSAE at ICLR 2026 — adapting the rotation trick to TopK sparse autoencoders, achieving 100% dictionary utilization vs. the baseline's 47%.

Skills

ML · AI

PyTorchPyTorch Lightningscikit-learnNumPyPandasLlamaIndex

Backend · DevOps

FastAPIFlaskDockerPostgreSQLAWSPyTestGitHub Actions

Languages

PythonC++TypeScriptJavaScriptJavaSQL

Experience

Current · Feb 2026 → Present

Stealth AI Startup

ML Research Engineer

  • Investigating inductive bias effects of CNN architectures (ResNet, ConvNeXt) on SSL representation geometry — diagnosing alignment, uniformity, and dimensional collapse
  • Two-stage experimental pipeline on CIFAR/STL-10 and ImageNet with ResNet-50; developing loss regularization and augmentation interventions targeting SSL training failure modes
May 2025 – Mar 2026

Algoverse AI Research

Research Fellow · Remote

  • Designed distributed training infrastructure with PyTorch Lightning across 100+ model configurations, cutting experiment runtime by 30% through optimized data loading and gradient checkpointing
  • Built experiment tracking and evaluation framework processing 10K+ runs, enabling systematic ablations across hyperparameter sweeps to surface statistically reliable results
  • Adapted the rotation trick to TopK sparse autoencoders via custom PyTorch autograd functions, achieving 100% dictionary utilization vs. baseline's 47% and 6.1× lower feature overlap

University of Washington

Paul G. Allen School of Computer Science & Engineering

Bachelor of Science in Computer Science

Statistical Machine LearningAlgorithm Design & ComplexityAbstract Linear AlgebraReal AnalysisDiscrete Mathematics
Expected June 2028

4.0 GPA

Dean's List

Projects

Equivariant VAE for Video Generation

PyTorchPyTorch LightningFastAPIDockerAWS

Designed a VAE enforcing equivariance in the latent space to ensure geometric transformations in input frames correspond to predictable latent trajectories, improving temporal consistency in generated video.

SVD & Eigendecomposition Engine

NumPyPython

Derived and implemented a full numerical linear algebra engine in pure NumPy — Gram-Schmidt QR factorization, power iteration with deflation, and two-sided Jacobi SVD. Validated against scikit-learn with residuals under 10⁻¹⁰.

Achievements & Involvement

Achievements

  • ·NSF REU — AFUS Program, NC State · Autonomous systems, edge-assisted cooperative perception for CAVs under Dr. Wujie Wen (May – Jul 2026)
  • ·Kaggle Bronze Medal — Medical Image Segmentation · Top 100 teams (top 7%), 0.87 Dice score on UW-Madison MRI segmentation
  • ·Dean's List · 4.0 GPA
  • ·Published at ICLR 2026 @ Re-Align Research Workshop

Involvement

  • ·ColorStack Member
  • ·NSBE — National Society of Black Engineers
  • ·CodePath Technical Interview Prep
  • ·Open Source Contributor