I'm part of the founding team of an AI Stealth Startup, where I work on the software and infrastructure developed. More details coming soon.
I am the Autonomous Division Lead for Formula Electric at Berkeley, where I lead the software, hardware, and electrical aspects of building a fully autonomous race car. This includes perception, environment mapping and planning, controls, drive-by-wire, electrical integration, and making it all work in real life. The team's goal is build the fastest, most reliable FSAE autonomous racing car in North America.
BS at UC Berkeley majoring in Computer Science and Electrical Engineering with a certification from the Sutardja Center for Entrepreneurship and Technology.
I was a software engineer and the youngest product manager at AI Camp at the time. Built out their course registration and user dashboard software that's processed 1,000+ users and $300,000+ in revenue since launch. Still currently in use.
I was captain of FTC Team 9614 Hyperion. Lots of CAD, manufacturing, and leadership lessons. Went to CA States a couple of times. 16th at the Maryland Tech Invitantional (defacto Worlds during COVID).
MESA is where I first discovered my interest in robotics and engineering. I participated in various challenges including developing tech for social equity, building Rube Goldberg machines, and designing egg drop mechanisms.
Notable achievements: Runners-up at the CA State competition for a prototype of an autonomous agricultural drone; 6th place finish at the CA State competition for designing a compact air quality system to combat CA wildfires.
Get to Know My Work.
Snake AI🔗
An AI trained to play snake, using a Deep Q-Network. Implemented fun little features like obstables and difficulty levels to test the AI.
Self Driving AI🔗
One of my first ML projects. I coded an feed-forward neural network to predict the steering angle of a car in simulation from scratch (no libraries), entirely in Javascript.
Explainstein🔗
Your own, video-based AI tutor! Built this with a few friends for a hackathon at Y Combinator. Ended up winning "Best Tech." Fun hack.
Sudoku Solver🔗
One of my first python projects where I used a backtracking algorithm to solve sudoku!
Model Predictive Control🔗
EDIT: MPC is a high-level control algorithm I implemented for the FEB. Used a bicycle model to model the system and various constraints to optimize the control inputs.
Drone Kahlo🔗
This is an autonomous drone that paints murals with spray paint. Designed and built the extraneous hardware to adopt the drone for this purpose. Includes a decent chunk of a fully autonomous pipeline, including perception of a defined canvas, PID-controlled hovering, grid path planning, and IMU-based state estimation. Most difficult part of this project was controlling the non-uniform body of mass of the drone.
Miscelleneous ML Projects
As a part of my Intro to ML class at Berkeley, I made many, mini ML projects, implementing concepts from support-vector machines and GDA to decision trees, neural nets, and PCA to predict MNIST digits, spam emails, titanic survivors, real-life locations of images, and more using well-known public datasets.
NFL Schedule Maker🔗
EDIT: Optimizes the nfl schedule for a given calendar year, taking into account the number of games each team plays, the number of games each team plays at home, and the number of games each team plays on the road.