Aarush Aitha Email & Phone Number

Dublin, CA, US

Berkeley, California, United States

Developing RL (reinforcement learning) algorithms for autonomous drone control. Integrating heuristic-based curriculum learning in combination with off-the-shelf model-free reinforcement learning algorithms such as proximal policy optimization (PPO), soft actor-critic (SAC), and deep deterministic policy gradient (DDPG).

Berkeley, California, United States

Developed software and built hardware for a blink-controlled RC car using EOG data streamed via Bluetooth from a Muse S headset and processed by Python code. Currently working on a project involving using ML to classify separable actions using EEG data.

Santa Clara, California, United States

Implementing a distributed systems architecture to increase deep learning simulation job throughput by 20%. Designing a new interface for specifying tags and knobs for GPU/model configurations to run LLM workloads. Investigating effects of different hybrid mappings and parallelisms on run time, memory usage, and token throughput

Berkeley, California, United States

Reinforcement learning for NLP, specifically program synthesis tasks such as code autocompletion and code optimization.

Berkeley, California, United States

Creating rubrics for student assignments and examinations. Grading student work. Assisting students one-on-one during office hours.

Berkeley, California, United States

Assisted students during discussion by providing hints and explanations when necessary. Taught problems to roughly 20 students relating to modular arithmetic, graph theory, and probability.

Berkeley, California, United States

Working in Professor Brian Barsky's lab on project 183C, Fingertip Keyboard Input. Using computer vision techniques to capture finger tracings of characters and classify them using convolutional neural networks. Created a confidence selection feature so that users can manually select ambiguous characters when the model fails to differentiate similar.

Pleasanton, California, United States

Created Splunk reports that monitored various types of server errors and compiled them into a dashboard.Led the development of dashboards to monitor front-end errors and performance metrics.

Santa Barbara, California, United States

Developed a new feature in order to generate a CSV report of important company information by querying PostgreSQL. Tested fixes on QA environments, which were then pushed and released to production after rigorous test cases. Fixed 15 complicated bugs involving front-end issues, back-end database problems, API calls, and more.

Completed an introductory project-based JavaScript and Firebase curriculum taught by Google engineers. Delivered a collaborative final project presentation that included a live demonstration to Google employees and community leaders.

Palo Alto, California, United States

Conducted research on the COVID-19 virus by analyzing roughly 4000 genome reads taken from the time frame of December 2019 to April 2020 (wide variety of locations around the world). Published a paper that drew conclusions about mutation frequencies by quantifying differences between the genomes. Continued work with mentors to test new k-mer analysis.

San Ramon

As a data science intern, I learned how to access and modify text-based databases such as HappyBase, MySQL, and graphical databases such as Neo4j using Python. For specific clients, I was able to automate and log recommendations. I also used RegEx to detect dates and byte locations in large files containing data about insurance claims.

Master Of Science - Ms, Electrical Engineering And Computer Science

Bachelor Of Science - Bs, Electrical Engineering And Computer Science, 3.92

Activities and Societies: Neurotech @ Berkeley, Tau Beta Pi, Eta Kappa Nu Relevant Coursework: Math W53: Multivariable Calculus Physics.

Advanced Scholar Diploma, Computer Science

Activities and Societies: VEX Robotics, AFA CyberPatriot, Environmental Technology Club GPA: 4.48/4