Muhammad Fateh Mujtaba

Machine Learning Engineer focused on building production AI systems across Generative AI, LLM applications, machine learning, computer vision, NLP, and AI backend engineering.

My work spans the full AI product lifecycle from data preparation, experimentation, fine-tuning, and evaluation to APIs, databases, integrations, deployment, observability, and production reliability.

I have built LLM applications, RAG systems, AI agents, recommendation systems, customer segmentation pipelines, text-to-SQL assistants, document intelligence workflows, OCR pipelines, multimodal AI systems, and real-time knowledge/data ingestion solutions.

On the backend side, I work extensively with Python and FastAPI, building REST APIs, async services, WebSockets, webhooks, background workflows, authentication-aware systems, persistent state, and external integrations. I have hands-on experience with PostgreSQL, MongoDB, Redis, event-driven architecture, event sourcing, concurrency control, state synchronization, retries, recovery, and production API workflows.

My GenAI stack includes OpenAI, Anthropic Claude, Hugging Face, LLaMA, Qwen, Cohere, Together AI, Azure OpenAI, LangChain, LangGraph, LlamaIndex, Ollama, Transformers, RAG, embeddings, semantic search, vector databases, reranking, structured outputs, tool calling, agent memory, prompt engineering, LoRA, QLoRA, and PEFT.

My broader engineering toolkit includes PyTorch, TensorFlow, Scikit-learn, Keras, pandas, NumPy, OpenCV, PaddleOCR, PyMuPDF, FAISS, ChromaDB, pgvector, Docker, Docker Compose, Git, GitHub, Playwright, Microsoft Graph API, Azure AI Document Intelligence, Azure Bot Service, Azure App Service, Azure, and AWS.

I’m especially interested in AI systems where models, data pipelines, APIs, databases, state, integrations, evaluation, and reliability all have to work together in production.