Saqlain Umer

Machine Learning Engineer

Codistan • Full-time

Aug 2023 - Present • 2 yrs 10 mos

I'm an AI Engineer focusing strongly on Generative AI and Retrieval-Augmented Generation (RAG) systems. With hands-on experience building production-ready AI solutions, I specialize in deploying LLM-based architectures that combine natural language understanding with efficient vector search. My expertise spans across: 1. GenAI & RAG pipelines using LLaMA, OpenAI, and LangChain 2. Vector databases: ChromaDB, Qdrant, Elasticsearch 3. Embedding generation, semantic search, and hybrid retrieval (dense + sparse) 4. Backend architecture using Python, FastAPI, and WebSockets 5. Scalable deployment with Docker, Docker Compose, PM2, and AWS EC2 6. Model optimization, including fine-tuning, quantization, and memory-efficient serving I’m passionate about solving real-world problems through AI — from intelligent document search engines to LLM-powered assistants. I work across the entire stack: from data ingestion and vectorization to inference, caching, and user feedback loops.