Final Year Student (8th Sem) @ COMSATS University Islamabad
Engineering SEEVIA: A multimodal AI ecosystem for real-time scene understanding and adaptive indoor mapping.
I am developing an end-to-end assistive framework focusing on two core domains:
- Perception Engine (CV/NLP): Real-time CNN-based object detection and contextual OCR for automated personal inventory and safety monitoring (Fall Detection).
- Autonomous Navigation (RL): Implementing Deep Reinforcement Learning to enable agents to generalize and pathfind across dynamic, unmapped indoor environments.
| Domain | Technologies & Architectures |
|---|---|
| Computer Vision (CV) | CNNs, YOLOv8, OCR (Tesseract/Google Vision), Image Segmentation |
| Machine Learning (ML) | Reinforcement Learning (DQN), Supervised Learning, Predictive Modeling |
| Natural Language (NLP) | Intent Recognition, Speech-to-Text (STT), Text-to-Speech (TTS) |
| Engineering / Ops | Python, PyTorch, TensorFlow Lite, FastAPI, Firebase, Git |
| Mobile Development | React Native, Expo (Cross-platform AI deployment) |
| Category | Status & Technical Focus |
|---|---|
| Main Framework |   |
| Research Niche |   |
| Deployment |   |
- Global Exposure: Alumnus of the Ajman University Student Exchange program (UAE).
- Interests: High-performance Edge AI, Multimodal Data Fusion, and Generalization beyond fixed supervised settings.