Hi, my name is
AI & Automation Developer
I build RAG systems, LLM-powered agents, and automated workflows that eliminate manual bottlenecks and help businesses work smarter.
01 · About
I help businesses work smarter through intelligent, AI-powered solutions. I specialize in designing and deploying RAG systems, LLM-based agents, and automated workflows that eliminate manual bottlenecks, reduce operational costs, and accelerate decision-making.
Backed by a strong foundation in Python and FastAPI, I build reliable, scalable systems tailored to unique business needs — from multi-tenant SaaS chatbot platforms to multi-agent booking assistants.
BS Information Technology
Government College University Faisalabad
2021 — 2025 · Faisalabad, Pakistan
Certificates
02 · Experience
May 2026 — Present
Lahore, Pakistan
Sep 2025 — Mar 2026
Faisalabad, Pakistan
03 · Projects
A multi-tenant SaaS RAG chatbot platform with role-based access control, JWT authentication, and tenant-isolated vector storage. Built as a React/FastAPI AI platform using LangGraph and Gemini embeddings with real-time streaming, document ingestion, analytics, conversation persistence, and cloud deployment.
An authenticated chat API with threaded, persistent conversations backed by LangGraph checkpoints. Supports PDF/TXT/Markdown upload with background processing into Pinecone for RAG retrieval, real-time streaming over Server-Sent Events, JWT auth, and email verification.
An AI rental-property search assistant that answers natural-language queries with live MongoDB data. An agent named “Smith” — built on the OpenAI Agent framework with Gemini 2.5 Flash Lite — filters by location, budget, bedrooms and property type, with relevance scoring and session-based conversation context.
A multi-agent airline booking API where Triage, Booking, and FAQ agents intelligently route user requests. Handles booking, updating, cancelling, and reviewing reservations with full passenger details, persisted in MongoDB, powered by Gemini through OpenAI-compatible APIs.
A RAG document assistant that lets users upload a PDF and query its content in natural language. Features session-based conversations with history, in-memory semantic indexing, per-IP rate limiting, and upload guardrails — built on FastAPI and Google Generative AI.
An end-to-end Retrieval-Augmented Generation service that indexes PDF documents into Pinecone and answers questions with Gemini. Returns both the generated answer and the retrieved source chunks (file, page, and content metadata), with per-document fingerprinting to skip re-indexing unchanged files.
04 · Skills
05 · Contact
I'm open to freelance projects, full-time roles, and collaborations around AI, RAG systems, and automation. Drop a message and I'll get back to you as soon as I can.