Anupam Kumar

Machine Learning Engineer with 4+ years of experience building production-grade conversational AI, agentic workflows, and Generative
AI systems. Skilled in orchestrating multi-step LLM pipelines with LangChain, designing RAG architectures with FAISS vector search, and optimizing
prompt engineering for accuracy, latency, and token efficiency. Built and deployed a production GenAI chatbot (Birbal) handling real business queries
via REST APIs with context management and >95% uptime. Proven track record in fine-tuning transformer models (LLaMA, BERT), containerizing and
deploying scalable real-time inference services with Docker and FastAPI on AWS, and automating CI/CD and monitoring for AI pipelines.