Hafeez Iqbal

I’m a Senior AI/ML Engineer with over 8 years of experience moving complex machine learning models from research environments into high-scale production. I focus on designing MLOps frameworks, automated pipelines, and distributed infrastructure that keep systems reliable and cost-effective.

I specialize in the Generative AI transition—optimizing Large Language Models (LLMs) with retrieval-augmented generation (RAG) and model quantization. In my recent role, I built an automated MLOps framework with MLflow that streamlined model promotion from R&D experimentation to production deployment.

I also deliver performance at scale: I developed a high performance RAG pipeline using Pinecone and LLMs to improve accuracy for complex domain queries. I optimized large-scale transformer models for real-time use with TensorRT and quantization to significantly reduce inference latency, and I scaled training to billion-parameter levels using distributed protocols on multi-GPU clusters with DeepSpeed.

To keep models healthy long-term, I establish rigorous monitoring and observability standards to detect data drift and maintain production integrity. Previously, I architected automated ingestion and augmentation pipelines, built internal performance dashboards with Flask and Streamlit, and operationalized systems with Docker, CI/CD, Kubernetes, and large-scale Spark workflows—from training to deployment to iteration.