The build log.

A command-line Tic Tac Toe game with object-oriented design featuring multiple game modes and intelligent AI strategies. Includes both random and strategic AI players that evaluate the board to win, block, or create opportunities.

A lightweight, from-scratch reimplementation of C++'s std::string class using only core C++ features. Demonstrates advanced concepts: Rule of Five, operator overloading, dynamic memory management, move semantics, and exception safety.

Professional console-based C++ receipt/memo system demonstrating real-world OOP patterns. Features dynamic data entry, formatted output, input validation, and Makefile build system. Separates concerns into helper classes: Item, Date, Address, CellNumber, Name, ShopDetails, ReceiptNo.

WhatsApp-based farming assistant using Gemini Vision AI for real-time crop disease detection, treatment plans, and weather alerts. Bilingual Urdu/English.

Answers questions from academic documents with cited sources. Built for students to interact with lecture PDFs and research papers.

Handles text, images, and voice messages. Intelligently routes to the right handler and responds in the correct modality.

Intelligent assistant framework using OpenAI Agents SDK with Chainlit UI. Features a main routing agent that intelligently dispatches queries to specialized sub-agents: Shaitani Calculator (humorous math), Web Search (real-time), and custom tools. Demonstrates multi-agent orchestration patterns.

A platform of specialized AI employees you can hire on demand — research, ops, content, and code agents collaborating via ACP.

Production WhatsApp agent with multi-intent routing, context memory, and human escalation paths.

Explored the impact of Fixed vs. Flexible Class Schedules on Student Productivity, Academic Success, and Satisfaction. Conducted surveys of 185+ students (aged 18–23), performed independent samples t-tests, KDE analysis, and statistical hypothesis testing to uncover relationships between schedule types and key outcome variables.

Comprehensive hands-on learning journey through the OpenAI Agents SDK. Covers agents, runners, results, streaming, tools, handoffs, lifecycle hooks, exception handling, guardrails, and multi-agent orchestration. Includes real-world projects: Fantasy World Generator and University Helpdesk Orchestrator.

Robust web scraping system for collecting book data from books.toscrape.com using AI-powered extraction with crawl4ai and Gemini LLM. Produces structured datasets (JSON & CSV) with 993 books including title, rating, price, availability, and cover images. Built for RAG applications and data science projects.

Sophisticated n8n workflow creating an intelligent WhatsApp bot with multi-modal capabilities. Processes text, voice, and images using Google Gemini AI. Features smart routing for message type detection, real-time voice transcription, image analysis, and conversation memory management. Fully integrated with WhatsApp Business API.

Minimal RAG pipeline transforming YouTube videos into intelligent Q&A systems. Extracts transcripts, builds vector embeddings with Google's embedding-001 model, and indexes with FAISS. Uses LangChain for orchestration and Gemini 1.5 Flash for answer generation. Jupyter-based, Colab-friendly, and Docker-ready.

Modern, responsive chatbot built with Streamlit and LangChain, powered by Google's Gemini AI. Features real-time streaming responses, 40+ model selection (Gemini 1.5, 2.0, 2.5), secure API key management, and a clean professional interface. Includes chat history tracking, dark/light theme toggle, and mobile-friendly responsive design.

Streamlit web application for intelligent document querying using Retrieval-Augmented Generation (RAG). Upload PDF or TXT files and ask natural language questions. Powered by Google Gemini embeddings, LangChain orchestration, and FAISS vector search. Includes multi-model support, chat history, and Docker deployment.

RESTful API built with FastAPI and Pydantic for managing patient records with automatic BMI calculation and health assessment. Features complete CRUD operations, data validation, sorting capabilities, and interactive Swagger/Redoc documentation. Demonstrates API design, error handling, and clean code practices.

Multi-agent workflow that autonomously researches, plans, writes, and formats professional newsletters. Uses Tavily for trending news research, Gemini 2.5 Pro for three specialized agents (Planning, Writing, Editing), and n8n for workflow orchestration. Generates weekly newsletters with proper citations, HTML formatting, and Gmail draft integration.

Multi-agent AI system for Pakistani farmers combining instant crop disease diagnosis via Gemini Vision, localized treatment plans, real-time supplier marketplace (50km radius), weather-based recommendations, and bilingual Urdu/English support. Built with Firebase Genkit, Next.js, and 6 specialized AI agents. Won Innovista Agentic AI Hackathon.

Intelligent chat agent built with OpenAI Agent Builder and ChatKit UI featuring a Router Agent that intelligently decides between weather queries (with widget output) and general conversation. Demonstrates structured output, conditional If/Else logic, and advanced agentic workflows with task routing.

Fully AI-generated interactive textbook on humanoid robotics and physical AI built using Spec-Kit Plus spec-driven development. Features 87 structured lessons, RAG-powered tutoring chatbot, FastAPI backend, Docusaurus frontend, Better Auth authentication, Qdrant vector embeddings, and adaptive content delivery. Zero manual coding - 100% generated through specifications. Includes CI/CD via GitHub Actions and deployments to Vercel & GitHub Pages.

Comprehensive Jupyter notebook data analysis project analyzing 30 Pakistan Cricket Team players. Processes player performance metrics including batting averages, strike rates, fitness levels, and role-based analysis using pure Python - no external libraries. Features 20+ analytical questions covering statistical analysis, fitness risk assessment, performance ranking, and team balance verification. Educational project demonstrating fundamental data science concepts.

Deep-dive statistical analysis of 215,000+ GitHub repositories to identify patterns driving open-source popularity. Discovered critical insights including forks as superior predictor (75%+ correlation), extreme data skewness (88.33), multicollinearity traps, programming language impact, and the dominance of open licenses. Comprehensive exploratory data analysis combined with correlation studies and machine learning modeling to reveal hidden patterns in open-source success.

Interactive Streamlit web application predicting GitHub repository popularity (Stars) using machine learning. Trained and compared 6 ML models (Linear, Ridge, Lasso, Random Forest, Gradient Boosting, XGBoost) on 215,000+ repositories. Features comprehensive exploratory data analysis, real-time predictions with scenario-based inputs, and deployed on Streamlit Cloud. Discovered critical insights: JavaScript/Python receive 3× more stars, MIT licensing shows 40% higher engagement, optimal repo size is 10-100 MB, active issue management correlates with 60% more forks.

Production-ready full-stack task management system demonstrating specification-driven development methodology. Phase II completion featuring Next.js 16 frontend with React 19, TypeScript 5, Tailwind CSS 4 glassmorphism UI, and FastAPI backend with PostgreSQL (Neon Serverless). 100% AI-generated using Claude Code and Spec-Kit Plus - not a single line manually written. Features secure JWT authentication, complete CRUD operations, task filtering, profile management, modern stats dashboard, and responsive design. Follows constitution-first approach with clean architecture patterns.

Intelligent esports coaching chatbot powered by GROQ's Llama 3.3 70B LLM. Provides personalized gaming strategy advice across 9+ games (Valorant, League of Legends, CS2, Fortnite, Apex Legends, Dota 2, Overwatch 2, Rocket League, Tekken 7). Features 4 dynamic coaching personalities: Competitive Pro Coach for rank climbing, Casual Fun Guide for enjoyable gameplay, Educational Analyst for technical learning, and Hype Man for motivational support. Advanced features include real-time streaming responses, adjustable detail level (1-10 slider), quick action buttons, and gaming-themed UI with glassmorphism design. Built with Gradio and deployed on Hugging Face Spaces.

Advanced agent-orchestrated retrieval-augmented generation (Agentic RAG) chatbot that intelligently reasons over user queries, decides when to retrieve document context, and generates grounded answers. Unlike traditional RAG systems that blindly retrieve for every query, this agent uses an LLM-driven decision engine to selectively invoke document retrieval tools. Supports multiple document formats (PDF, TXT, DOCX) with semantic chunking, vector similarity search using FAISS and Sentence-Transformer embeddings, and maintains conversational memory. Features source-aware answers with page numbers, voice-enabled AI (STT + TTS), query analytics, and real-time streaming responses. Built with OpenAI Agents SDK, LangChain, and Groq LLMs.

Comprehensive Power BI dashboard analyzing 4 years of retail sales data (80K+ records, 2022–2025) with advanced DAX time intelligence capabilities. Implements sophisticated time-based calculations including Year-over-Year growth analysis, Month-over-Month changes, Year-To-Date (YTD) metrics, Quarter-To-Date (QTD) performance, and Month-To-Date (MTD) tracking. Features multi-dimensional analysis across channels (Marketplace, Online, Retail, Wholesale), product categories (Electronics, Furniture, Fashion, etc.), and geographic regions (North, South, East, West). Interactive drill-down capabilities enable executive-level insights from high-level KPIs to granular monthly comparisons. Includes budget variance analysis, forecasting trends, regional performance heatmaps, and top-category growth rankings. Built with enterprise-grade data modeling and optimized DAX formulas for real-time performance monitoring.

Enterprise-grade 6-page Power BI dashboard analyzing TechMart's global sales performance across Jan–Jun 2023. Implements sophisticated multi-page architecture with specialized views: Comparison page for regional and team benchmarking, Trend page for time-series analysis and category contribution insights, KPI page for executive-level metrics with MoM tracking, Detail page for granular drill-downs into country-product matrices, Advanced page featuring salesperson efficiency analysis with outlier detection and anomaly identification, and Contribution page for team attribution. Features interactive visualizations including horizontal bar charts, column charts, pie charts, scatter plots, and waterfall charts. Includes advanced analytics like outlier detection (high ships, low sales) and salesperson profile drill-through capabilities. Data-driven insights on sales trends, team performance, regional variance, and product profitability. Built with optimized DAX measures and calculated columns for performance across 210K+ transactions.

Autonomous AI agent system that manages real personal and business workflows. Monitors multiple communication channels (Gmail, WhatsApp, LinkedIn) using watcher scripts and converts incoming events into structured tasks automatically. Features a local-first vault workflow that prioritizes and routes tasks with human-in-the-loop approvals for sensitive actions. Supports MCP-based external action architecture enabling flexible integration with various tools and services. Provides comprehensive observability with live dashboard showing watcher health, queue state, and activity history. Implements autonomous execution loop with weekly CEO briefing reports. Silver tier complete with Gold-tier integrations in progress. Demonstrates deep hands-on understanding of AI automation, reliability, observability, and safe agent design patterns.

Comprehensive gig economy platform built to bring financial transparency, accountability, and empowerment to millions of gig workers in Pakistan. Addresses critical gaps including lack of unified income records, hidden deductions, and no way for workers to prove earnings to banks or landlords. FairGig provides three distinct role-based dashboards: Worker platform for logging shifts across platforms (Careem, Bykea, Foodpanda) with screenshot verification, Verifier dashboard for reviewing and validating submissions, and Advocate analytics for monitoring commission trends and detecting worker vulnerability. Core features include Z-score anomaly detection to flag unusual deductions with human-readable insights, privacy-first city-level analytics without exposing individual worker data, and print-ready income certificates for real-world use with banks and institutions. Built with 6 microservices architecture (FastAPI + Node.js with shared PostgreSQL database) designed for production-style scalability under hackathon constraints. One-command startup system enables judges to fully run the platform in seconds.