Resume

Software Engineer Intern

DoorDash

Architected and implemented an identity verification dedupe system for DoorDash’s internal Fraud Workstation Platform. Serviced 240,000 dashers with 100 write QPS and 50ms latency, decreasing Average Handling Time by 3 minutes. Constructed UI components in React, BFF endpoints in Typescript, and Kafka event consumers & gRPC endpoints in Kotlin to dynamically persist user verification statuses in a PostgreSQL database.

• Kotlin, gRPC, Typescript, React, PostgreSQL

Software Engineer Intern

DoorDash

Migrated internal Django monolithic entity tooling to microservice-based architecture, serving 200,000 daily requests. Developed GraphQL BFF endpoints in Typescript, and built React front-end components to dynamically auto-generate and validate form schemas for flexibly searching, editing, creating, and deleting arbitrary database entities. Expanded backend authorization for CRUD endpoints for 14,000 support users and integrated Elasticsearch indexing in Kotlin.

• React, Typescript, Kotlin, gRPC, GraphQL

Research Assistant

The Cottrell Lab - SMART 4.0

Identified and validated different architectures and methods on improving molecular fingerprint classification accuracy and F1-score from Nuclear Magnetic Resonance and Mass spectra data using PyTorch Lightning. Improved upon previous CNN-based architectures through transformers, feature engineering, and positional encoding of spectra data in terms of cosine similarity and informational-retrieval metrics, obtaining a Recall@1 of 0.9. Prototyped architectures using multimodal (CLIP) and language modeling (BART) techniques to generatively predict molecular SMILES strings and obtain latent representations of spectral data.

• PyTorch, PyTorch Lightning

Machine Learning Programmer

Scripps Institution of Oceanography

Experimented with image processing techniques and Vision Transformers to enhance plankton image classification, achieving an accuracy of 87.66%. Constructed and deployed web applications in Plotly and Dash to visualize experiment results, inspect mis-classified images and confusion matrices, and efficiently compare 100,000 individual images within plankton datasets.

• PyTorch, Plotly, Dash

Software Engineer Intern

Bentley Systems

Developed front end for live-sharing map application, integrating layout managers and integrating with other Bentley Systems software.

• C#, .NET, Javascript

Masters of Science

Carnegie Mellon University

Machine Learning

• Advanced Introduction to Machine Learning
• Advanced NLP
• Probability and Mathematical Statistics

Bachelors of Science

University of California, San Diego

Computer Science

• Deep Learning
• Recommender Systems
• Computer Vision
• Probabilistic Reasoning
• Statistical Natural Language Processing
• Reinforcement Learning
• Convex Optimization
• Numerical Analysis
• Probability and Stochastic Processes
• Computer Security
• Computer Networks
• Operating Systems

Summa Cum Laude

University of California, San Diego

Graduated within the top 2% of the senior class