Computer Science Engineer specializing in AI/ML, Computer Vision, and Generative AI — I build systems that actually ship, from real-time edge inference to production RAG platforms.
-
🔭 I'm currently working on: An Edge AI Computer Vision SDK that runs fully in-browser (YOLO26n + ONNX Runtime Web), enabling real-time object/face detection with zero installation across any device.
-
👯 I'm looking to collaborate on: Machine Learning and Computer Vision projects — especially real-time detection systems, edge AI deployment, and applied Generative AI / RAG pipelines.
-
🤝 I'm looking for help with: Optimizing model performance for browser environments (WebAssembly, Web Workers) and squeezing out lower-latency real-time inference.
-
🌱 I'm currently learning: Advanced deployment patterns for cross-platform ML — model quantization, ONNX optimization, and scaling RAG systems in production.
-
💬 Ask me about: Python, TensorFlow, Computer Vision (OpenCV, YOLO, MediaPipe), NLP/LLMs, RAG architectures (Qdrant, RAGAS, Langfuse), and building ML systems that run anywhere.
-
⚡ Fun fact: I enjoy building ML systems that work everywhere — from laptops to mobile browsers — with zero setup required. 🚀
- Edge AI Computer Vision SDK — YOLO26n trained on 14,449 images (97.1% mAP@50), deployed via ONNX Runtime Web for real-time browser-based proctoring/detection.
- AI-Powered Gesture Recognition — Real-time 21-landmark hand tracking (MediaPipe/OpenCV) powering air drawing, presentation control, and system control apps.
- Clinical Assistance RAG Platform — Production-grade medical QA system with Qdrant, BM25 + RRF hybrid search, CrossEncoder reranking, and RAGAS evaluation.