Generative AI, NLP & Computer Vision Engineer
AI Engineer specializing in Generative AI, NLP, and Computer Vision. Expert in Python, PyTorch, YOLOv11, and LangChain, with a proven track record of building production-ready, real-time AI systems that power impactful solutions across education, healthcare, and enterprise domains.
Passionate about building transformative solutions that create real impact
Currently a final-year undergraduate in Artificial Intelligence (UMT), I've engineered end-to-end deep learning and full-stack solutions—ranging from custom YOLOv11 deployments for real-time detection, to retrieval-augmented (RAG) lecture agents and memory-driven conversational bots.
In every project, my focus is production-quality code, clear ML explainability, and tangible value for users and businesses alike. My leadership includes guiding student teams, conducting workshops, and mentoring on advanced computer vision, NLP, and workflow automation.
When I’m not building or deploying systems, I contribute to open source, share insights on Hugging Face, Kaggle, and GitHub, or document my AI journey to help the next generation innovate.
Writing scalable, production-ready AI systems with an emphasis on maintainability and real-world impact.
Staying ahead of technology trends and actively exploring new research to push the boundaries of what's possible.
Working with teams, mentoring peers, and building clear documentation to multiply collective expertise.
Optimizing workflows, automating repetitive tasks, and ensuring seamless user experiences through high-performance solutions.
Technologies I've worked with in real-world projects and professional environments
From enterprise solutions to tech innovation, building scalable systems and leading AI-driven initiatives across diverse technology stacks
Selected as a Software Development Intern(~20–25 hours/week), focusing primarily on .NET technologies. Involved in core software projects requiring rapid adaptation to new tech stacks, timely delivery of high-quality code, and effective team collaboration within a professional learning environment.
Contributing to ongoing AI/ML research with a focus on object tracking, real-time detection, and advanced YOLO architectures. Actively working toward robust multi-object tracking solutions for surveillance and industrial contexts, as part of a final-year project. Engaged in early experimentation, integration, and methodology documentation for future publication.
Improved database operations and automated data reporting for vocational education programs. Enhanced event management workflows and technical troubleshooting.
Assisted the course instructor by managing LMS activities, organizing and uploading class materials, and overseeing student assignment and project submissions.
Designed and delivered workshops on Generative AI, large language models, and prompt engineering for early-career tech professionals and students.
Spearheaded development of a reservation web system for event space management including admin dashboards, user portals, and automated e-mail notifications.
AI-driven systems for vision, language, and automation—integrated for real-world education, industry, and innovation.
Real-Time Unified Lecture Extraction Network
AI-powered multilingual educational assistant automating lecture transcription (99+ languages), bilingual notes in Roman Urdu/Hindi & English, and curriculum-aligned quiz creation with RAG. Supports dashboards, CLO/PLO mapping, multi-format upload, and integrates with learning outcomes for institution readiness.
Memory-Augmented Conversational Agent
A rule-based agent simulating five human-like memory systems with Neo4j graph storage, real-time dashboards, web chat, and ESP32 hardware for multimodal input and voice. Small step toward AGI and real-time agent reasoning.
Multi-Pipeline Face/Gender/Celebrity Recognition
CV pipeline using YOLOv11 for face detection, EfficientNetV2-S for celebrity recognition (15+ Pakistani celebrities). 1st place in CV Kaggle competition; real-time inference, visualizes bounding boxes and recognizes gender, class imbalance handled.
Elderly Safety & Surveillance
YOLOv11-based fall detection system for real-time video feeds in safety/public spaces. Fine-tuned on LE2I, high precision/recall, supports logging and alerts, modular for sensor integration.
Event Space Reservation Web App
Designed and developed a full-stack .NET and SQL Server application for event reservations. Features include user registration, an admin dashboard, automated email notifications, and Excel report generation. Built as a proof-of-concept to demonstrate robust reservation workflows in a simulated environment.
IoT-Driven Attendance Platform
Designed and built a prototype automated attendance system using ESP32 with NFC/fingerprint authentication, a speech-to-text pipeline via Flask, and an OLED display. Developed as a semester term project to demonstrate end-to-end integration of IoT hardware, biometric verification, and real-time data processing. Focused on core features, system workflow, and proof-of-concept usability in a classroom scenario.
Interested in collaborating on innovative AI/ML projects?
Research contributions and technical insights
Research on real-time fall detection using YOLOv11 and computer vision.
An in-depth explanation of Principal Component Analysis (PCA), its importance for dimensionality reduction, and its value in machine learning workflows. Covers how PCA works, why it matters, and practical implementation tips.
A practical guide to choosing the right statistical imputation method for missing data. Explains when to use mean, median, or mode, including their pros, cons, and suitable scenarios in data preprocessing.
Explores the application of 1D Convolutional Neural Networks in handling sequential data, covering their working principles, advantages over traditional methods, and key use cases. Includes a Python TensorFlow/Keras implementation.
A comprehensive overview of gradient descent, the fundamental optimization algorithm behind most machine learning and deep learning models. Discusses the mechanics, variations, and significance of gradient descent in achieving accurate predictions.
An in-depth guide to LSTM networks, their role in handling sequential data for deep learning tasks like NLP and forecasting, and their advantages over basic RNNs. Includes explanations of working mechanisms and core applications.
Covers Gated Recurrent Units (GRUs) as an efficient alternative to LSTMs for sequential data in deep learning. Explains their architecture, how they function, their advantages, and practical applications.
Honoring accomplishments in AI research, academic excellence, and impactful technical leadership.
School of Systems & Technology, UMT
Awarded by Dean for outstanding academic performance and top CGPA in Artificial Intelligence program.
Lahore Garrison Education System
Secured the Gold Medal at the Matric level for academic excellence.
Ready to bring your ideas to life? Let's discuss your next project