I build full-stack and AI-powered software, with a strong interest in backend systems, distributed systems, and practical LLM applications.
Currently pursuing my MS in Computer Science at Santa Clara University and building projects around AI agents, secure APIs, and cloud-native systems.
- AI tools that turn messy real-world workflows into simple user experiences
- Backend systems with clean APIs, reliability, and security in mind
- Full-stack products that are actually usable, not just demos
- Distributed systems, middleware, and cloud-native applications
Voice-driven AI assistant for completing complex PDF forms through natural conversation.
Built to help users fill government-style documents without manually reading and typing through every field.
What it does
- Understands PDF forms using a Vision-Language Model
- Asks users contextual questions through voice
- Validates answers before filling fields
- Writes responses directly into the PDF
Built with: React, FastAPI, Qwen2.5-VL, vLLM, Whisper, ElevenLabs, PyMuPDF
🏆 Winner — Most Likely to be a Unicorn, Hack for Humanity
Zero Trust API Gateway for securing distributed microservices.
Built as a backend/security project focused on service identity, encrypted communication, and resilience patterns.
What it includes
- Continuous authentication principles inspired by Zero Trust architecture
- TLS 1.3 and mTLS-based service verification
- Circuit breakers, exponential backoff, and concurrency limiting
- FastAPI services behind an NGINX gateway
Built with: Python, FastAPI, NGINX, TLS/mTLS, Docker
- 🦄 Most Likely to be a Unicorn — Hack for Humanity
- 🥇 1st Place — Sentient Cohort 0 Arena Challenge 0 (The Big Q)
Languages: Python, JavaScript, C++, SQL
Frontend: React, Redux
Backend: FastAPI, Node.js, REST APIs
AI/ML: LLMs, RAG, LangChain, Whisper, Vision-Language Models
Tools: Docker, AWS, Firebase