Muhammad B, B is an AI Engineer specializing in building scalable, production-grade AI agent systems and LLM-powered applications. He has strong expertise in orchestrating agentic workflows using frameworks like LangChain, LangGraph, and the A2A protocol, alongside implementing robust LLMOps and MLOps practices with tools such as Arize Phoenix and LangFuse. He is highly experienced in developing Retrieval-Augmented Generation (RAG) pipelines, real-time AI systems including ASR with speaker diarization, and microservices-based backend architectures using FastAPI, Flask, gRPC, and WebSockets. His cloud expertise spans AWS and Azure, where he deploys containerized applications using Docker, Kubernetes (Helm), and CI/CD pipelines with AWS CodePipeline, leveraging services like OpenSearch, Neptune, and ECR. With a solid foundation in Python and C++, he has also worked across embedded systems, robotics (ROS 2), and UAV control systems, demonstrating versatility across both software and hardware domains. Muhammad combines strong system design skills with hands-on experience in vector databases such as Qdrant and Chroma, real-time communication protocols, and modern AI tooling, enabling him to deliver high-performance, scalable, and intelligent solutions across diverse applications.