I design and deploy AI systems — from fine-tuned LLMs and RAG pipelines to production MLOps infrastructure — at the intersection of data science and cloud engineering.
Designed and deployed production-grade GenAI solutions leveraging LLMs, RAG architectures, and vector databases. Built end-to-end ML pipelines from data ingestion through model serving, with full MLOps lifecycle management on cloud infrastructure.
Developed NLP and computer vision models for enterprise clients. Automated feature engineering, model training, and evaluation workflows. Collaborated with data engineers to build scalable data pipelines for model training and inference.
Built dashboards, ran A/B tests, and developed statistical models to support product decisions. Established data quality processes and wrote Python automation scripts that reduced manual reporting effort by 60%.
A retrieval-augmented generation system that ingests enterprise documents and answers questions with source citations. Uses semantic chunking, hybrid search (dense + sparse), and a re-ranker for high-accuracy responses.
An automated pipeline for fine-tuning open-source LLMs (Llama, Mistral) on domain-specific data using LoRA/QLoRA. Tracks experiments with MLflow, auto-evaluates with ROUGE and BERTScore, and deploys to cloud endpoints.
An end-to-end ML platform with automated data preprocessing, model training, and real-time prediction serving. Features a clean Streamlit UI and scheduled retraining triggered by data drift detection.
Open to AI/ML engineering conversations, research collaborations, and interesting data science challenges. Reach out — I respond fast.
