Austin Narcomey
PhD Candidate
New Haven, Connecticut, United States
Summary
👤
Senior🎓
Top SchoolAustin Narcomey is a PhD candidate at Yale researching personalized robot behavior and human-AI interaction, with eight years of experience spanning robotics, computer vision, and real-time systems. He built counterfactual explanations for robot decision-making and now focuses on integrating human feedback to adapt agents to individual users. A Stanford BS/MS alum with strong academic performance, he has hands-on research experience from Stanford's CV and HCI labs and practical industry internships at Amazon Lab126 and Northrop Grumman. His open-source work includes contributing vision assignments to Stanford's widely used CS131 course—implementing classic detectors, descriptors, and robust matching pipelines—demonstrating both pedagogical and engineering fluency. Based in New Haven, he blends rigorous research with production-minded software engineering to push explainable, personalized autonomy toward real-world use.
8 years of coding experience
3 years of employment as a software developer
Doctor of Philosophy - PhD, Robotics and Artificial Intelligence, Doctor of Philosophy - PhD, Robotics and Artificial Intelligence at Yale UniversityBachelor of Science (BS), Master of Science (MS), Computer Science, BS GPA: 3.87, MS GPA: 4.03, Bachelor of Science (BS), Master of Science (MS), Computer Science, BS GPA: 3.87, MS GPA: 4.03 at Stanford University
Spanish, English
Github Skills (4)
computer-vision10
python10
ransac10
scikit-image9
Programming languages (3)
Github contributions (5)
StanfordVL/CS131_release
Oct 2020 - Nov 2020
Released assignments for the Stanford's CS131 course on Computer Vision.
Role in this project:
ML Engineer
ML EngineerContributions:9 commits, 9 pushes in 1 month
Contributions summary:Austin contributed to the release of assignments for a Computer Vision course. The commits detail the implementation of a Harris corner detector, descriptor matching techniques, and the use of RANSAC for robust feature matching. They also worked on implementing the HOG descriptor and constructing image pyramids for object detection, ultimately building panoramic images.
stanfordpythonvisiondeep-learningcomputer-vision
ANarcomey/visdial-rl
Oct 2019 - Nov 2019
PyTorch code for Learning Cooperative Visual Dialog Agents using Deep Reinforcement Learning
Contributions:17 pushes in 23 days
pytorchvisual-dialogdialogstable-baselinesdeep-learning
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.