David

PhD Candidate

Available for 2024 Roles

Building the future of AI with robust & scalable vision systems.

I am a PhD candidate specializing in Computer Vision and Machine Learning. My research focuses on self-supervised learning, generative models, and their applications in medical imaging and autonomous systems.

Previously, I interned at Google DeepMind and Meta AI. I am passionate about open-source and making AI accessible to everyone.

campaignLatest Updates

Jun 2024

Accepted to CVPR 2024

My paper on "Self-Supervised Learning for Small Datasets" was accepted for an oral presentation.

Apr 2024

Guest Lecturer at Stanford

Invited to speak about the future of Generative AI in Healthcare diagnostics.


Self-Supervised Vision Transformers for Medical Image Segmentation

David, Sarah Connor, John Doe

CVPR 2024starOral Presentation

Generative Adversarial Networks for Climate Modeling

Jane Smith, David

ICCV 2023

Efficient Neural Architecture Search

David, Alan Turing

NeurIPS 2022

Selected Projects


Experience & Education

downloadDownload Resume

PhD Student, Computer Science

2021 - Present

Massachusetts Institute of Technology (MIT)

Researching self-supervised learning methods for medical imaging. Advised by Prof. Brilliant Mind. TA for "Intro to Deep Learning".

Research Scientist Intern

Summer 2023

Google DeepMind

Developed a novel transformer architecture for video understanding, improving SOTA on Kinetics-400 by 1.5%.

AI Research Intern

Summer 2022

Meta AI (FAIR)

Contributed to PyTorch core library optimizations for distributed training.

B.S. in Computer Science

2017 - 2021

Stanford University

Graduated with Honors. Focused on Artificial Intelligence and Human-Computer Interaction.