I'm an AI Engineer and ML Specialist with a background in Statistics and Mathematics, specialized in building end-to-end AI systems — from model design and fine-tuning to production deployment, observability, and continuous optimization.
My experience spans RAG and agentic architectures end-to-end: data ingestion and preprocessing, chunking and embeddings, vector storage, semantic retrieval, reranking, prompt orchestration, multi-agent workflows, evaluation, monitoring, and scalable deployment. I've built these systems using LangChain and LangGraph as core orchestration layers, with OpenSearch, Pinecone, Chroma, and Redis for retrieval and state management, and with AWS, GCP, and Azure to support cloud-native AI infrastructure.
I place strong emphasis on production reliability and measurable impact, using tools such as MLflow, Phoenix Arize, CloudWatch, Docker, Kubernetes, FastAPI, and GitHub Actions to track performance, evaluate quality, monitor failures, and ship maintainable AI systems with proper CI/CD and observability.
Beyond LLMs, I bring deep experience in traditional machine learning — forecasting, probabilistic scoring, NLP, fine-tuning, and predictive analytics with Prophet, ARIMA, LSTM, BERT, T5, and GPT-based architectures. That gives me the judgment to choose the right approach for each problem, whether generative, predictive, or a hybrid system that combines both.
My background in RLHF, annotation, and model evaluation helps me approach AI from a quality and alignment perspective, not just a systems perspective. I can help organizations design, build, deploy, and improve AI products end-to-end — from early architecture decisions to reliable production systems that support automation, decision-making, and user-facing intelligence.