Software Engineer specializing in Generative AI and LLMs with a proven track record of building production-ready automation systems. Expert in LangChain, RAG, and FastAPI, I've delivered high-impact solutions ranging from real-time driver monitoring systems to AI-powered telepsychology platforms. Passionate about bridging the gap between cutting-edge AI research and scalable business applications.
Lahore, Pakistan
Experience
WORK
Software Engineer (AI Automation)
Freelance / Independent Contractor
Designed and deployed a production-grade WhatsApp AI automation system for a paying client.
•Real-time message routing and session management
•LLM response generation via LangChain and REST APIs
•Containerization and deployment using Docker
Key Achievements
→Automated business workflows, reducing manual response workload by 80% for a paying client.
→Optimized LLM response accuracy through iterative prompt engineering and fine-tuning, ensuring reliable business-grade interactions.
→Built an end-to-end pipeline with automated query classification and context-aware responses using LangChain.
→Implemented structured logging to PostgreSQL and session management to support concurrent multi-user conversations without state collision.
Teaching Assistant - Introduction to Computing, Basic Electronics
Forman Christian College University
Evaluated quizzes and assignments and provided constructive feedback to students. Supported 60+ students across sessions, resolving queries and reporting progress to the instructor.
•Evaluated quizzes and assignments and provided constructive feedback to students.
•Supported 60+ students across sessions, resolving queries and reporting progress to the instructor.
Sep 2025 – 2026
WORK
Teaching Assistant - Computer Networks
Forman Christian College University
Supported 25+ students with networking concepts including TCP/IP, OSI model, and routing protocols across weekly sessions. Assisted in grading assignments and maintaining academic records for a 15-week semester course.
•Supported 25+ students with networking concepts including TCP/IP, OSI model, and routing protocols across weekly sessions.
•Assisted in grading assignments and maintaining academic records for a 15-week semester course.
TCP/IPOSI modelRouting protocols
Project Portfolio
PERSONAL
Present
Real-Time Driver Drowsiness Detection System
Developer
Developed a 4-stage CV pipeline using MediaPipe and MobileNetV2 for real-time driver monitoring, processing video at 20+ FPS on CPU with end-to-end latency under 50ms. Engineered a dual-model architecture for Eye/Yawn classification, achieving 98.65% validation accuracy using two-phase fine-tuning. Implemented temporal decision logic with a 60-frame sliding window, reducing false-positive alert rate by filtering single-frame anomalies before triggering driver alerts.
Built a secure telepsychology platform using Next.js for the frontend and FastAPI for backend services, containerized with Docker and backed by Supabase/PostgreSQL. Integrated Google Gemini API through LangChain to analyze projective drawings and PHQ-9/DASS-21 questionnaires, reducing assessment time by 20–30%. Developed a scalable matchmaking microservice in Python targeting a 35% improvement in therapeutic alignment by dynamically pairing patients with psychologists based on specialisation, availability, and assessment outputs.
Next.jsSupabasePostgreSQLGemini APILangChainDockerFastAPIAPI Development