I build AI-native systems end-to-end β from the infrastructure that doesn't flinch under load, to the LLM layer that makes products intelligent, to the interfaces users actually ship with. I don't just integrate AI as a feature; I architect around it. Finishing my final semester at ASU and open to full-time roles starting May 2026.
- π₯ Winner: 1st Place β Hacks for Humanity 2025
- π Winner: 3rd Place β HackASU 2025 (DeFi Track)
- π₯ Winner: 2nd Place β Innovation Hacks 2.0 @ ASU (AntonRx Track)
- π Built: RxRefactor β two-tier RAG pipeline (SQLite + ChromaDB/Claude) replacing hours of manual payer PDF research with sourced answers
- π Built: Dementia Care β native Android app with Vapi-powered voice MMSE assessments and a PubMedBERT/pgvector RAG backend on Gemini 2.5 Flash Lite
- π Built: E.V.O β Next.js + Wagmi DeFi options dApp on Ether.fi
- π‘ Currently exploring: AI infrastructure, LLM orchestration, low-level systems & RTOS
- Interests: LLM Orchestration Β· RAG Systems Β· Distributed Systems Β· Consensus Algorithms (Raft) Β· DeFi Β· Embedded Systems Β· RTOS
| Project | Description | Stack |
|---|---|---|
| RxRefactor | Two-tier drug policy intelligence platform β SQLite structured lookups + Claude RAG over ~6,100 ChromaDB chunks. Innovation Hacks 2.0 β 2nd Place (AntonRx Track) | Next.js, FastAPI, Claude, ChromaDB, PyMuPDF |
| E.V.O β DeFi Options dApp | DeFi options platform on Ether.fi β HackASU 3rd Place | Next.js, Wagmi, Viem, RainbowKit, Solidity, Hardhat |
| Dementia Care App | Native Android app with Vapi-powered voice MMSE assessments and a PubMedBERT/pgvector RAG backend on Gemini 2.5 Flash Lite | Kotlin, Jetpack Compose, Flask, pgvector, LangChain, Vapi AI |
| CS2 Telemetry Pipeline | 20GB/hr real-time match telemetry with zero data loss | Go, Kafka, CockroachDB, Prometheus |
| Cloud-Native Orchestration Platform | Stateless ephemeral envs scaling to 4,000+ users | Java, Docker, MongoDB, Vue.js |