Hanan Zahoor — AI/ML Engineer and Full-Stack Developer

// what he built

• Built an LLM-powered AQI advisor using Python, GPT-4o, Agno, Firecrawl, and Gradio.
• Designed an agentic workflow using prompt engineering techniques to retrieve live AQI data and generate personalized recommendations from structured and unstructured sources.
• Integrated GPT-4o APIs to generate context-aware health recommendations from AQI, weather, medical history, and user activity — improving response quality through iterative prompt refinement and evaluation.
• Integrated live REST APIs with validation and error handling to improve recommendation reliability.

// concepts

LLM applications · AI agents · Live data retrieval · Structured & unstructured data · Prompt engineering · Model evaluation

// what he built

• Built a Linux log analyzer using Python, Flask, React, and MongoDB for real-time monitoring and debugging.
• Engineered an ML pipeline using TF-IDF, LSA, and ensemble learning to classify known attacks and detect anomalous system behavior.
• Developed Flask REST APIs and a real-time dashboard for log ingestion, visualization, model inference, and security alert management.

// concepts

Machine learning pipeline · Text classification · Anomaly detection · Log ingestion · REST APIs · Real-time dashboard · Security monitoring

// what he built

• Built a privacy-first social platform using React, TypeScript, Express.js, and SQLite with secure real-time communication.
• Implemented AES-GCM encryption and PBKDF2 key derivation for end-to-end encrypted messaging.
• Developed disappearing messages, encrypted file sharing, QR-based key exchange, and secure REST APIs.

// concepts

Full-stack development · Encryption · Secure communication · Key exchange · Backend APIs · Privacy-focused product design