I build AI-powered tools that help businesses search, understand, and act on their own data — chatbots, semantic search, and document intelligence systems, backed by solid full-stack engineering underneath.

📍 Based in Pakistan · Open to freelance & contract work 💼 LinkedIn · 📧 [email protected]

I work across the full stack, but the throughline is turning unstructured or messy data into something people can actually use — whether that's a chatbot answering questions from a document set, a dashboard surfacing trends from raw data, or a backend that ties a bunch of moving parts together reliably.

• 🤖 AI & LLM applications — RAG systems, semantic search, chatbots grounded in real data (not just prompting a model and hoping)
• 🔧 Full-stack development — Flask, FastAPI, and clean backend architecture
• 📊 Data analysis & dashboards — turning raw operational data into decisions
• 🧩 Practical engineering — I care more about whether something works reliably on real data than whether the demo looks flashy

AI assistant that answers medical questions using semantic vector search over a document knowledge base (Pinecone + NLP). Built to retrieve relevant information rather than rely on the model's raw memory — grounded answers, not guesses.

A chat interface for querying messy real-world technical documentation — mixed PDFs, markdown, and wikis — with source citations on every answer and a custom evaluation suite to measure retrieval and answer accuracy.

Interactive Power BI dashboard analyzing hotel booking patterns, cancellations, and guest behavior from raw operational data — Python + Pandas for the data pipeline underneath.

A recommendation system with a working backend, built to surface personalized results from user preference data.

E-learning platform teaching Pakistan Sign Language through video lectures, quizzes, and progress tracking — full product build, not just a script.

Languages: Python · JavaScript · PHP · Java AI/ML: Pinecone · Scikit-learn · TensorFlow · OpenCV · LLM prompt engineering Backend: Flask · FastAPI Data: Pandas · NumPy · Power BI Other: Git · MySQL · MongoDB

• 3 projects submitted across multiple hackathons (online and in-person)
• 🥈 X Hackathons Level 2: submitted separate projects to separate hackathons
• 🔀 Dual Boot: competed in both in-person and online hackathons
• 🧭 Generalist: submitted to hackathons with 3 different themes
• 🎯 First in-person hackathon · First online hackathon

👉 View all achievements on Devpost

• Advanced RAG architecture & retrieval evaluation
• Agentic workflows and tool-calling with LLMs
• CI/CD pipelines with GitHub Actions
• Real-time APIs & WebSocket integrations

If you've got messy data, documents, or workflows that need an AI layer on top — I'd like to hear about it. Open to freelance projects and collaborations.

⭐ If any of this is useful to you, a star or a follow is appreciated.