umair Amjad

I’m a Senior AI Engineer with 5+ years of experience designing and deploying production-grade AI systems, multi-agent architectures, and scalable backend APIs using Python. My focus is building reliable LLM-based solutions that perform under real-world load.

I’ve designed and deployed 8 multi-agent AI systems using LangChain and Observability with LangGraph. I built FastAPI services handling 10K50K daily requests with stable performance, and reduced runtime errors by approximately 30% through structured Pydantic validation.

To keep systems dependable, I integrated Langfuse observability, improving debugging efficiency by approximately 40%. I also built LLM agent workflows for automation, reasoning, and task orchestration, collaborating with teams to deliver production releases within deadlines.

Before that, I worked as a Machine Learning Engineer developing 5 ML models (accuracy ranging from 82% to 90%) and processing datasets of 50K to 300K records. I delivered ML APIs serving 1K to 5K daily requests, improved pipeline efficiency by approximately 22%, and implemented monitoring for model performance and drift.