Open to research internships · Summer 2027

AlexChen

I build machine learning systems that areinterpretableandhonestly reliablein the open world.

Ph.D. Candidate @MIT CSAIL ·  Cambridge, MA

Portrait of Alex Chen
Trustworthy ML Group
02

News

  1. One paper accepted to NeurIPS 2026 as a Spotlight 🎉
  2. Paper on certified robustness accepted to ICML 2026.
  3. Started a research internship at Google DeepMind (Mountain View).
  4. Gave a talk on sparse circuits at the MIT ML Tea — slides here.
  5. Our ICLR paper on disentangling LLM features was selected for an oral presentation (top 1.8%).
  6. Received the Example Fellowship for research on trustworthy AI.
03

Research Interests

01 · Interpretability

Interpretable Deep Learning

Reverse-engineering the circuits and representations inside transformers — turning black boxes into systems we can inspect, edit, and trust.

Mechanistic InterpretabilityProbingCircuits
02 · Robustness

Robustness & Uncertainty

Making models reliable under distribution shift, with calibrated uncertainty and guarantees that survive contact with the real world.

OOD GeneralizationConformal PredictionCalibration
03 · Efficiency

Efficient Foundation Models

Long-context attention and sparse computation — cutting the cost of foundation models without cutting corners on quality.

Long ContextSparse AttentionDistillation
04

Selected Publications

2026

Alex Chen, Maya Zhang, Priya Patel, Wei Lin

NeurIPS 2026#Interpretability#Theory

Alex Chen, Sofia Ramirez, Jane Doe

ICML 2026#Robustness#Uncertainty
2025
2024

Alex Chen, Wei Lin, Omar Hassan, Jane Doe

arXiv preprint#Efficiency#Long Context
arXivCode
05

Experience

Education

  1. 2023 — Present

    Ph.D. in Computer Science

    MIT CSAIL

    Advised by Prof. Jane Doe · Trustworthy ML Group

  2. 2019 — 2023

    B.S. in Mathematics & Computer Science

    Example University

    Summa cum laude · Thesis on sparse attention

Experience

  1. Summer 2026

    Research Scientist Intern

    Google DeepMind

    Interpretability team — scaling circuit analysis to production-scale models.

  2. Summer 2025

    Research Intern

    Meta FAIR

    Conformal methods for open-world recognition.

  3. 2021 — 2023

    Undergraduate Researcher

    Example Lab

    First encounters with attention sparsification.

06

Contact

Say hello

Let's build something trustworthytogether.

The fastest way to reach me is email — I usually reply within 48 hours. Always happy to chat about research, internships, or collaboration.