I'm Muhammad Khubaib, an AI Engineer based in Islamabad, Pakistan. I have 2+ years of AI engineering experience, focused on Generative AI, LLMs, and agentic systems.
I architect and deliver production-grade RAG pipelines, multi-agent systems, and enterprise AI platforms on Azure — turning business needs into scalable, secure, cost-efficient solutions with measurable operational impact.
Junior AI Engineer — Mobiz IT (Islamabad, Pakistan) · May 2025–Present
Building enterprise AI for banks, ministries, and US clients: agentic RAG chatbots (Microsoft Agent Framework, Databricks Mosaic AI Vector Search), regulatory compliance scoring against SAMA-CSF and NCA-ECC, 5-agent orchestration on Azure AI Foundry, multimodal RAG ingestion validated with RAGAS, and multi-agent platforms on the Claude Agent SDK with Langfuse and Helicone observability. Also built an AI recruiter that cut initial screening time by 80–85%.
AI Trainee — Mobiz IT (Islamabad, Pakistan) · Jan 2025–Apr 2025
Built internal AI apps (adaptive skills assessment, interview question generator) and an enterprise RAG ingestion pipeline with FastAPI, Azure Document Intelligence, Azure AI Search, and Azure Functions.
BS Artificial Intelligence — COMSATS University Islamabad (CUI), Islamabad Campus · 2021–2025
Final GPA: 3.53 / 4.00
- Databricks Certified: Generative AI Engineer Associate (2025)
- AWS Certified: AI Practitioner (2025)
- Microsoft Certified: Azure AI Engineer Associate (2025)
Generative AI & LLMs: LLMs, NLP, multi-agent & agentic systems, RAG, hybrid & semantic search, multimodal RAG, GraphRAG, structured-first retrieval, chunking strategies, model evaluation, prompt engineering, grounding & guardrails, Agent2Agent, Model Context Protocol, LLM observability (Langfuse, Helicone).
Frameworks: Microsoft Agent Framework, Claude Agent SDK, LangGraph, LangChain, Microsoft 365 Agents SDK, Copilot Studio, OpenAI API, PyTorch, FastAPI, SQLAlchemy, Alembic, Pydantic.
Cloud & data: Azure (AI Foundry, OpenAI, App Service, Document Intelligence, Logic Apps, Event Grid, Functions), Databricks (Asset Bundles, Unity Catalog, Mosaic AI Vector Search), Microsoft Graph API, Docker, CI/CD, PostgreSQL, MySQL, Azure AI Search, Neo4j, FalkorDB.
