🎓 MS in Data Science, UMD · 🤖 Agentic LLM systems · ⚙️ GPU kernels to production pipelines
I build systems that go from data → model → deployment → real-world use. MS in Data Science from UMD (GPA 3.97), with prior experience as a Technical Analyst at Oracle and recent work building agentic AI systems at MTech Ventures.
- Agentic LLM Systems — MCP servers, tool orchestration, multi-agent reasoning loops
- LLM Fine-tuning & Inference — QLoRA, vLLM, multi-GPU deployment
- RAG Pipelines — retrieval systems with eval gates and citation grounding
- CUDA & GPU Systems — kernel optimization, memory hierarchy, edge GPU benchmarking
- Generative Models — diffusion models, segmentation-guided synthesis
- MLOps & Infrastructure — Docker, Kubernetes, CI/CD for ML
- An agentic RAG system in production use, with eval-gated CI/CD and autoscaling
- An MCP tool server used by analysts at MTech Ventures
- CUDA kernel benchmarks with roofline analysis on edge GPUs
- LLM fine-tuning and multi-GPU inference pipelines
- Generative and applied NLP projects outside the core infra work
- ML / AI — Python, PyTorch, HuggingFace, vLLM, PEFT/QLoRA
- Systems — CUDA, C++, Docker, Kubernetes, AWS, GitHub Actions
- Data — SQL, PySpark, Pandas, Pinecone, MLflow
