About
Experiences
- Quant Trading @ Jane Street (NYC, Summer 2022): Attended classes, participated in mock-trading, and worked on data analysis projects with trading desks. Further details omitted due to IP restrictions.
- Software Engineering @ Sentry.io (Remote, Fall 2021): Worked with the Data Team based in San Francisco building ETL pipelines for 100M+ daily messages from the Sentry.io platform. Used Apache Kafka, PySpark, GCS, AirFlow, BigQuery, and DataDog to prototype a batch processing pipeline with increased throughput and scalability.
- Quant Trading @ Jane Street (NYC, Summer 2021): Attended classes, participated in mock-trading, and worked on two data analysis projects with trading desks. Further details omitted due to IP restrictions.
- ML Researcher @ Yale IMG Lab (Yale, 2020-21): Built a deep learning pipeline analyzing social group phenomena for human-robot interaction using graph neural networks, improving metrics over state-of-the-art heuristic baselines. Paper available here.
- Software Engineering @ Gatherly (Remote, Fall 2020): Spearheaded data analytics initiative analyzing 1M+ user actions to extract actionable insights for online event hosts. Implemented an automated payment system using the Stripe API with subscriptions, discounts, and tiered pricing. Used AWS, NodeJS, ReactJS, and Stripe.
- Data Analyst @ RTI International (Remote, Summer 2020): Designed campus monitoring dashboard to visualize temporal building ocupancy. Implemented automated demand prioritization system for BI team, reducing manual labor by 80%. Used Microsoft SQL Server, Power BI, and Excel.
- ML Researcher @ Yale LILY Lab (Yale, Fall 2019): Trained machine learning model to classify salient events in 2D physics simulations with 91% accuracy. Contributed to research paper studying implicit physical reasoning in NLP. Paper available here.
Projects
GitHub Trends dives deep into the GitHub API to bring you insightful metrics on your contributions, broken by repository and language.
Key Stat: 2000+ users have created a GitHub Wrapped
Statbotics modernizes FRC data analytics by calculating and disseminating key metrics including Elo and component OPRs
Key Stat: Over 40,000 users have used Statbotics
Hopscotch helps you discover suitable neighborhoods based on many factors: budget, grocery stores, gyms, subway lines, bike share, parks, safety, proximity to specific locations and more
Key Fact: Built Hopscotch in just one week at HackLodge.
Extending existing GNN explanation algorithms for node and graph tasks to link prediction on social network and user-object graphs. Contributing to the open-source PyTorch Geometric library.
Key Fact: Working with Prof. Rex Ying at Yale
Uses LLMs and Levenshtein distance to suggest corrections for typos and mental mistakes. Integrated with VSCode suggest directly in the IDE.
Key Fact: Built at YHack Mini 2022, 3rd Place
Clockwork takes your deadlines, events, and goals, and creates a balanced and healthy schedule, factoring in personal preferences.
Key Fact: Built at YHack 2022, 2nd Place
Quickbites integrated with Uber Eats, Doordash, and GrubHub to find the cheapest and fastest food delivery options based on your location, requirements, and preferences.
Key Fact: Downloaded over 1000 times on the App Store
Implemented similarity, sentiment, and topic analysis on Coursetable reviews using NLP techniques. Wrote a Medium article highlighting key Python libraries to build upon.