Aditya Shukla

I’m an AI/ML engineer and LLMOps professional with end-to-end experience across the GenAI lifecycle—from RAG pipeline design and prompt engineering to LLM evaluation and production-oriented deployment.

I build and operationalize LLM-powered applications using Python with LangChain and FastAPI, grounded by vector search (FAISS) and Hugging Face embeddings for document ingestion, chunking, indexing, and retrieval. I focus on making systems reliable in production by improving output quality—accuracy, hallucination reduction, and instruction-following.

In my recent work evaluating enterprise-grade model families, I monitored LLM output quality and used LLMOps-style practices (reproducibility, benchmarking, and scalable evaluation frameworks) to support faster, higher-confidence model releases.

I also design automation around prompt engineering and evaluation pipelines, and I’ve built agentic systems (using LangChain) for tasks like personalized cold email generation. I’m comfortable working across the stack—RAG and agentic workflows, evaluation, guardrails/alignment concepts, and containerized deployment with Docker and Kubernetes.