Crafting Intelligence, One Model at a Time
I'm a passionate Artificial Intelligence student at FAST NUCES, Karachi. My journey in tech revolves around building intelligent systems that bridge the gap between theoretical AI and real-world applications.
From developing hybrid resume-job matching systems using LLMs to creating agentic AI workflows with LangGraph, I thrive on pushing the boundaries of what's possible with machine learning and natural language processing.
As a Teaching Assistant, I've had the privilege of mentoring students across courses ranging from Programming Fundamentals to Artificial Intelligence, deepening my own understanding while helping others grow.
Technologies I Work With
Languages
Frameworks
Tools
Platforms
Soft Skills
Academic Background
FAST National University (NUCES)
Bachelor of Science — Artificial Intelligence
📍 Karachi, Pakistan
Relevant Coursework
Certifications
Cloud Computing Fundamentals
IBM
Master in Data Science, Data Analytics & Data Analysis
Udemy
Microsoft Learn Student Ambassadors (MLSA)
FAST Karachi Chapter — Event management and community involvement
Professional Journey
Agentic AI Intern
Wellness Innovations • Remote
Teaching Assistant
FAST NUCES • Karachi, Pakistan
Featured Projects
Resume–Job Matching AI
Full-stack Flask app that uses LLMs to semantically match resumes to job descriptions. Produces a detailed compatibility report with skill gap analysis, match score, and structured JSON output. Deployed on Vercel.
Chat with Your Document
Fully local RAG chatbot powered by LLaMA 2 + LangChain. Upload any PDF and have a natural language conversation with it — no API keys, no cloud. Uses FAISS vector store and HuggingFace embeddings for semantic retrieval.
DevOps Monitoring Dashboard
Production-grade observability stack with Flask, Prometheus, Grafana, and Pushgateway. Collects 30+ real-time metrics every 5 seconds across 4 professional dashboards covering CI/CD, app health, and system performance.
APR Dataset Builder
Automated pipeline that mines GitHub repositories to build structured datasets of buggy/fixed Python code pairs for training Automated Program Repair models. Resumable, incremental, and RepairLLaMA-compatible.
Hybrid Movie Recommender
Streamlit app combining user-based collaborative filtering (Pearson correlation) and content-based filtering (genre cosine similarity). Fetches live movie posters via OMDB API. Deployed on Streamlit Community Cloud.
Django E-commerce Platform
Full-stack laptop marketplace built with Django featuring AI-powered product recommendations, smart search, cart management, order tracking, and a complete admin panel for inventory control.
HEC Transcript Generator
Streamlit app that validates student transcripts against HEC Pakistan CS degree requirements. Auto-categorizes courses, calculates CGPA, checks 7 credit categories, and generates a downloadable PDF transcript.
IR Boolean Search Engine
Streamlit IR system implementing Boolean Retrieval and Proximity Search from scratch over a research abstract corpus. Supports AND/OR/NOT operators, Porter stemming, and pre-built inverted + positional indexes.
VSM Search Engine
Vector Space Model IR system with TF-IDF weighting and cosine similarity for ranked document retrieval. Built from scratch over research abstracts with a pre-built NumPy matrix index and Streamlit UI.
Video Game Recommender
Content-based filtering system for video games using genre overlap, developer matching, and TF-IDF text similarity on game summaries. Features tiered scoring, fuzzy title search, and Precision@K evaluation.
Let's Connect
I'm always open to discussing new opportunities, collaborations, or just having a chat about AI and technology.
Interested in working together?
Whether it's a research collaboration, project idea, or job opportunity I'd love to hear from you.
Send me an Email