👨‍💻 About Me I'm an AI Infrastructure / LLM Serving engineer and MS in Data Science candidate at CU Boulder (GPA 4.0/4.0). I focus on making LLM inference fast, reliable, and memory-efficient under load — benchmarking and fixing the points where KV cache, throughput, and latency break down on constrained GPUs. 🔭 Building KV-cache-aware LLM serving on vLLM (PagedAttention, preemption, prefix caching) 🛠️ Interested in inference optimization, GPU memory management, and serving-stack reliability 🏥 Ex-Stanford Health Care (AI Dev Intern) · 🏦 Ex-Oracle (2 yrs, core banking at scale) 🏆 Red Bull Innovation Track Winner — HackCU12

• KV cache & memory management — PagedAttention block pools, preemption/recompute, cache saturation
• Serving stacks — vLLM, OpenAI-compatible APIs, concurrent load testing, prefix caching, chunked prefill
• Performance eval — throughput, TTFT, p50/p99 latency, Prometheus /metrics instrumentation
• GPU & HPC — NVIDIA A100 (MIG), CUDA version matching, Slurm scheduling, Apptainer/Docker containers

vLLM · PagedAttention · CUDA · PyTorch · Slurm/HPC · Apptainer/Docker · Prometheus · FastAPI

LLMs & Agents: LangChain · LangGraph · RAG (FAISS + BM25 + RRF) · ReAct Agents · Pydantic · Multi-Agent Systems
ML & Data: PyTorch · TensorFlow · Scikit-learn · Pandas · NumPy · PySpark
Cloud & MLOps: AWS · GCP · Azure · Docker · Podman · Kafka · Airflow · Databricks · Snowflake
Languages: Python · Scala · C/C++ · PL/SQL · Shell/UNIX · SQL

🏥 AI Developer Intern — Stanford Health Care (May 2026 – Jul 2026 · 3 mo)

• Built a real-time BACnet/IP data pipeline and database from scratch for hospital telemetry feeding ML/agentic workflows
• Deployed on-premise LLM agents with tool-calling, async concurrency, and hardened error handling; owned LLM-serving-stack uptime and health monitoring in a regulated environment

🏦 Software Developer — Oracle (Financial Services) (Aug 2023 – Aug 2025 · 2 yrs)

• Owned core banking modules serving 600+ institutions (UBS, Access Bank); built data pipelines in Python/PL/SQL/Shell/REST
• Automated an ETL + validation framework (10K+ daily txns, defects −30%); ran containerized deployments (Docker, Podman) at 99.9% uptime

📊 Data Engineer Intern — The Thick Shake Factory (May 2022 – Jul 2022 · 3 mo)

• Built high-throughput ETL/ELT streaming topologies on Apache Kafka with parallel ingestion via Airflow DAGs, consolidating data from 40+ outlets into columnar storage
• Optimized cluster utilization to cut pipeline bottlenecks, powering real-time dashboards over 100K+ daily transactions