← Back to portfolio

Muhammad Abubakar

AI Engineer | Python Developer | LLM & RAG Systems
📍 Swabi, Pakistan — open to Islamabad & Lahore 📞 0340 5058254 ✉️ m.abubaakar755@gmail.com 🔗 /in/abubakar755 💻 ABUBAKAR755

Professional Summary

AI Engineer with hands-on experience building LLM-powered agents, retrieval-augmented generation (RAG) workflows, Python APIs, web-scraping systems, and production data pipelines. Comfortable taking AI features from prototype to deployment with LangGraph, LangChain, FastAPI, Docker, vector search, and modern data tooling. Seeking AI Engineer, Python Developer, Generative AI, or backend automation roles in Islamabad and Lahore.

Technical Skills

Languages & BackendPython, FastAPI, REST APIs, PostgreSQL, Redis AI & LLM SystemsLangGraph, LangChain, RAG, vector search, OpenAI API, Anthropic Claude, multi-agent orchestration Machine LearningScikit-learn, PyTorch, TensorFlow, deep learning, computer vision, data preprocessing Data & AutomationWeb scraping, Apify, ETL workflows, n8n, Google Maps API, data cleaning and validation Engineering ToolsDocker, Git, GitHub, deployment workflows

Professional Experience

AI EngineerDec 2025 — Present
SAT Software House (PSEB)
  • Engineer data scraping and processing pipelines across 10+ government departments using Python, Apify, and the Google Maps API.
  • Build automated ETL workflows that clean, structure, transform, and validate business-intelligence data for government and commercial use.
  • Design AI automation systems with LLMs, RAG pipelines, and vector search, supported by scalable FastAPI services and Docker deployment.
AI DeveloperSep 2025 — Nov 2025
ProspectROI AI
  • Architected and deployed 75+ autonomous AI agents with LangGraph and LangChain, including task decomposition and multi-agent orchestration.
  • Built an AI-driven sales automation platform reported to improve lead conversion by 40%, integrating GPT-4 and Anthropic Claude into production workflows.
  • Engineered web-mining and data-extraction systems processing 1M+ data points daily, alongside end-to-end machine-learning pipelines.

Selected Projects

LLM Sales Agent

Multi-agent LangGraph workflow for lead qualification and outreach across CRM, email, and chat channels using LangChain, FastAPI, GPT-4, and Anthropic Claude.

NexaBoost Agency Website

Responsive digital-agency website with Next.js 15, TypeScript, Tailwind CSS, and Framer Motion. Source code published on GitHub.

Support Reply Assistant

AI support-reply assistant built with React, TypeScript, Node.js, SQLite and Anthropic Claude — github.com/ABUBAKAR755/dafinitiq-support-reply-assistant

Education

BS Computer Science — University of Engineering and Technology, MardanCompleted Jul 2025

Certification

Machine Learning A–Z — UdemyJul 2025

Portfolio and code samples: github.com/ABUBAKAR755