Biography
I’m a 4th year Electrical Engineering and Computer Science at UC Berkeley. I currently do Systems and Machine Learning research at UC Berkeley RISE Labs under Joseph Gonzalez. With Moustafa Adelbaky, I’m researching serverless execution across mutli-cloud, 5G, and the edge. With Yaoqing Yang and Yaodong Yu, I’m working on machine learning distribution shift explainability. I am currently planning to go to 5th years at UC Berkeley.
- Parallel & Distributed Systems
- Machine Learning
- Operating Systems
- Optimization
-
Undergraduate, 2022
University of California, Berkeley
-
High School, 2018
Lynbrook High School
Work Experience
Investigating how overparameterization affects generalizations on spurious correlations
Design and run research experiments to test model size, subgroup robustness and counts, and pretraining
Automatically constructed spurious correlation large scale image datasets with Imagenet
Created an efficient serverless scheduling framework across client, 5G networks, and the cloud.
Wrote heuristic and linear programming based algorithms for optimal code placement on compute nodes
Optimized memory and cost with serverless Kubernetes, AWS, GCP, and Azure
Built paramater tracking to process models used in ranking content in the Alexa Mobile App
Designed and built dev tool for experiment tracking, managing A/B model testing, and dashboards
Supported popular open source features on AWS SAM-CLI(5.7K stars)
Added support for cloudformation and cloudformation intrinsic templating resolution in AWS SAM-CLI
Built a Serverless Testing Framework to benchmark and test AWS Serverless infrastructure
Projects
An efficient serverless scheduling framework across client, 5G networks, and the cloud. It uses heuristic and linear programming based algorithms for optimal code placement on compute nodes. It optimizes memory and cost with serverless Kubernetes, AWS, GCP, and Azure.
Current planning to publish to NSDI 2022. We are currently researching multi-cloud datastores and dag execution(Step functions across the cloud).
Prior work has suggested that overparameterization can hurt test accuracy on rare subgroups. Motivated by the fact that subgroup information is often unknown, we investigate the effect of model size on worst-group generalization under empirical risk minimization (ERM). Our systematic evaluation reveals that increasing model size does not hurt, and may help, worst-group test error under ERM. Published at NeurIPS DistShift Workshop 2021. Full Paper
3D Modeling and animation using Maya about a mouse trying to reach the moon. The video was edited using After Effects and Premiere Pro.
Created a Smart Speaker system to study passive listening devices.
Our work is open source and available at (https://github.com/blues-lab/passive-listening-prototype, https://github.com/blues-lab/passive-listening-classifier-training, https://github.com/blues-lab/passive-listening-classifiers)
- Built machine learning NLP models to classify different intents for passive listening and speech
- Setup architecture and workflow to run smart speaker with dynamic skills for experiment
- Wrote noise filtering/detection algorithms in order to improve and detect intents
A 3D Water Simulation using C++ and Navier Stokes Equations.
It uses optimized water physics using KD Trees and parallel processing
- Modeled car speed, accident rate, lane size, switching lanes, and many more to simulate the traffic
- Handled multithreaded events in golang synchronizing using mutex/locks
- Created realtime demo with websockets, cli, and react.js
Courses & Skills
Skills
Languages: Python, Golang, Java, C, Scala, Rust, Verilog, Risc-V/x86, C++, CUDAWeb: React, Flask, Django, Node.js, GraphQL, MySQL, Mongodb, Redis, Cassandra
Machine Learning: Pytorch, Tensorflow, Natural Language Processing, Keras, OpenCV
DevOps: Kubernetes, Docker, Jenkins, AWS, GCP, Azure, Serverless
Animation: Maya, Blender, After Effects, Premiere Pro
Game Development: Unity
EE/CS Upper-Division Coursework:
Systems
- CS 162: Operating Systems and System Programming
- CS W186: Introduction to Database Systems
- CS 161: Computer Security
- CS 164: Programming Languages and Compilers
Machine Learning
- CS W182: Designing, Visualizing and Understanding Deep Neural Networks
- CS 189: Introduction to Machine Learning
- EE 127: Optimization Models in Engineering
- EE 126: Probability and Random Processes
- CS 188: Introduction to Artificial Intelligence
- EECS 290: Hardware for Machine Learning
- CS 194-15: Special Topics - Parallel Programming
Signal Processing
- EE 123: Digital Signal Processing
- EE 120: Signals and Systems
- ME C134 / EE C128: Feedback Control Systems
Misc
- CS 170: Efficient Algorithms and Intractable Problems
- EECS 151: Digital Design and Integrated Circuits + FPGAs
- CS 184: Foundations of Computer Graphics
Hobbies
During my spare time, I enjoy reading comics and playing games. I also enjoy creating 3D animation and working on game development.