About
I am an AI/ML Engineer and Software Developer specializing in Generative AI, particularly Large Language Models and Retrieval Augmented Generation systems. I graduated as a Silver Medalist in Software Engineering from UET Taxila with a 3.94 CGPA.
My experience includes optimizing inference latency, building evaluation frameworks that reduce hallucinations, and containerizing data pipelines for production environments. I combine a strong theoretical foundation in Deep Learning and Computer Vision with hands-on engineering experience across the stack.
Skills
Experience
Software Engineer
- Built and managed full-stack promotional workflows, integrating third-party APIs with custom backend logic to automate multi-channel marketing campaigns.
- Containerized internal data pipelines and audience-tracking applications using Docker, ensuring consistent environments across development and staging.
- Analyzed user behavioral metrics and API data patterns to optimize internal targeting tools, directly improving digital engagement.
Frontend Developer
- Developed and deployed responsive web interfaces using HTML5, CSS3, and JavaScript, translating design mockups into pixel-perfect pages.
- Optimized client-side rendering and cross-browser compatibility, testing across multiple devices for a seamless user experience.
Digital Marketing Intern
- Led data-driven SEO and social media campaigns for cross-industry clients, improving search visibility and audience engagement.
- Managed end-to-end project lifecycles with cross-functional teams, delivering 100% of projects on time and within budget.
Projects
ThoraxNet
Production grade medical AI framework using BioMedCLIP (ViT-B/16) for multi-label chest X-ray pathology detection, with Monte Carlo Dropout for uncertainty quantification.
DeepGuard-XAI
Deepfake detection platform built on a fine-tuned EfficientNet-B4, with GradCAM explainability that shows exactly where a video was manipulated, plus an InsightFace pipeline.
ClinicaQuery-AI
RAG assistant for querying clinical papers with strict source attribution and confidence scoring, built to tame medical hallucinations in language models.
Devops-swarm
Autonomous AI developer that spins up a secure cloud sandbox, writes code, runs tests, and relentlessly self-corrects until the GitHub issue is solved. Built with LangGraph.
AgroVision-Net
Deep learning system for plant disease classification across 38 categories, fine-tuning EfficientNetB0 on the PlantVillage dataset with a Gradio web interface.
Ask-My-Portfolio RAG
The retrieval engine running inside this site. A TF-IDF index over my profile with cosine ranking, running fully client-side with no server.
Live Lab
Four experiments: two deployed models answering over the network, a retrieval engine running in your browser, and visualizations written by hand.
Education
2020–24
Bachelor of Science in Software Engineering
Graduated with distinction, ranking among the top of the class with consistent academic excellence across the program.