I build production-grade AI systems that turn unstructured data into usable products.
Over the past 10+ years, I’ve worked across ML, data infrastructure, and backend systems, with a focus on document intelligence and LLM-powered applications. My work includes building RAG pipelines, OCR processing systems, and API-driven ML services used in financial, retail, and enterprise environments.
Recently, I’ve been focused on:
• Designing RAG systems for search and question answering over large document datasets
• Building end-to-end pipelines (OCR → preprocessing → indexing → inference)
• Improving model accuracy through prompt design, evaluation workflows, and data iteration
• Deploying ML systems with monitoring, retraining, and performance optimization
I’m comfortable owning systems end-to-end — from experimentation to production — including infrastructure, APIs, and MLOps.
Tech: Python, FastAPI, PyTorch, Airflow, MLflow, Kubernetes, AWS, GCP, Azure
Open to roles focused on applied AI, LLM systems, and production ML platforms.
