About
I'm a junior at San Jose State University studying CS. I've had the opportunity to intern at leading tech companies including Amazon, Meta Research, and Cloudglue, where I've worked across various tech stacks from Full-Stack, Android, AI, Backend, and more. I'm a team player and value engineering excellence, customer obsession, and speed in my professional work.
Work Experience
Skills
Check out my latest work
I've worked on a variety of projects, from simple websites to complex web applications. Here are a few of my favorites.
Partner In Crime
Engineered a multi-modal intelligence pipeline capable of ingesting 300+ page PDFs and YouTube URLs to extract investigative entities using Llama Vision models. Visualized crime scene reconstructions by generating interactive 3D environments from 2D images and text descriptions using Three.js and WebGL. Automated the generation of chronological case timelines by vectorizing evidence into a RAG retrieval system to identify patterns across disparate data sources.
Disconnected Data Distribution
Architected a delay-tolerant network protocol to transmit internet data via physical infrastructure (e.g., transit systems) using standard Android devices and USB interfaces. Engineered an asynchronous backend in Spring Boot and gRPC to manage packet routing and synchronization between offline mobile nodes and a central MySQL database. Migrated the client-side architecture from Java to Kotlin and Jetpack Compose to improve UI performance, laying the groundwork for cross-platform support via Kotlin Multi-Platform (KMP).
Gloq
Built real-time network analysis and threat detection software implementing Tool-Former and Auto-GPT research concepts. Developed a cross-platform desktop app using Electron for client-side Python executables to monitor network activity. Parallelized LLM processes through LangGraph's Multi-Agent framework and utilized RPC to reduce API calls by 50%.
Ghost Guide
Developed an AI tutor that provides personalized assistance on user's desktop applications like Figma, Blender, CAD, etc. Decreased token costs 30% by implementing a segmentation algorithm with Gemini to position CSS of React components. Built a cross-platform Electron overlay to process YouTube API's text, audio, and video for multi-modal LLM analysis.
Meet Smarter
Developed a real-time video calling site by streaming voice-to-text segments into an Apache Kafka driven LLM pipeline. Deployed a SingleStore vector database with a RAG agent, achieving sub-50ms query speeds for efficient context retrieval.
I like building things
During my time in university, I attended 5+ hackathons. People from around the country would come together and build incredible things in 2-3 days. It was eye-opening to see the endless possibilities brought to life by a group of motivated and passionate individuals.
- C
CalHacks 12.0
Berkeley, CA
Engineered a multi-modal intelligence pipeline capable of ingesting 300+ page PDFs and YouTube URLs to extract investigative entities using Llama Vision models. Visualized crime scene reconstructions by generating interactive 3D environments from 2D images and text descriptions using Three.js and WebGL. Automated the generation of chronological case timelines by vectorizing evidence into a RAG retrieval system to identify patterns across disparate data sources. - L
LA Hacks
Los Angeles, CA
Developed an AI tutor that provides personalized assistance on user's desktop applications like Figma, Blender, CAD, etc. Decreased token costs 30% by implementing a segmentation algorithm with Gemini to position CSS of React components. Built a cross-platform Electron overlay to process YouTube API's text, audio, and video for multi-modal LLM analysis. - S
SF Hacks
San Francisco, CA
Built Fillosophy, a platform that processes videos with OpenAI vector embedding to support a RAG workflow to automate medical forms for immigrants. Implemented document agentic workflows to extract relevant fields of documents using LlamaIndex and Google Gemini. Won "Best Biotech Hack" & "People of Color Empowerment" tracks, placing 1st out of 300+ competitors. - C
Columbia DevFest 2025
New York, NY
Built Gloq, a real-time network analysis and threat detection software implementing Tool-Former and Auto-GPT research concepts. Developed a cross-platform desktop app using Electron for client-side Python executables to monitor network activity. Parallelized LLM processes through LangGraph's Multi-Agent framework and utilized RPC to reduce API calls by 50%. Won "Best General Project", placing 1st out of 250+ students. - C
CalHacks 11.0
San Francisco, CA
Developed a real-time video calling site by streaming voice-to-text segments into an Apache Kafka driven LLM pipeline. Deployed a SingleStore vector database with a RAG agent, achieving sub-50ms query speeds for efficient context retrieval.
Get in Touch
Want to chat? Feel free to reach out via email or LinkedIn. I'll respond whenever I can.