Waqas S.

𝐓𝐨𝐩 𝐑𝐚𝐭𝐞𝐝 𝐏𝐥𝐮𝐬 𝐅𝐫𝐞𝐞𝐥𝐚𝐧𝐜𝐞𝐫 | 𝟏𝟎𝟎% 𝐉𝐨𝐛 𝐒𝐮𝐜𝐜𝐞𝐬𝐬 𝐒𝐜𝐨𝐫𝐞 𝐏𝐡𝐃 𝐢𝐧 𝐂𝐨𝐦𝐩𝐮𝐭𝐞𝐫 𝐕𝐢𝐬𝐢𝐨𝐧 | 𝟏𝟓+ 𝐘𝐞𝐚𝐫𝐬 𝐢𝐧 𝐀𝐈, 𝐌𝐋 & 𝐂𝐨𝐦𝐩𝐮𝐭𝐞𝐫 𝐕𝐢𝐬𝐢𝐨𝐧 𝐎𝐛𝐣𝐞𝐜𝐭 𝐃𝐞𝐭𝐞𝐜𝐭𝐢𝐨𝐧, 𝐒𝐞𝐠𝐦𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧, 𝐓𝐫𝐚𝐜𝐤𝐢𝐧𝐠 & 𝐕𝐢𝐬𝐢𝐨𝐧 𝐅𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧 𝐌𝐨𝐝𝐞𝐥𝐬 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈 | 𝐕𝐢𝐬𝐢𝐨𝐧 𝐋𝐚𝐧𝐠𝐮𝐚𝐠𝐞 𝐌𝐨𝐝𝐞𝐥𝐬 | 𝐋𝐋𝐌𝐬 | 𝐀𝐠𝐞𝐧𝐭𝐢𝐜 𝐀𝐈 𝐑𝐓𝐒𝐏, 𝐂𝐂𝐓𝐕 & 𝐑𝐞𝐚𝐥 𝐓𝐢𝐦𝐞 𝐕𝐢𝐝𝐞𝐨 𝐀𝐧𝐚𝐥𝐲𝐭𝐢𝐜𝐬 𝐏𝐮𝐛𝐥𝐢𝐬𝐡𝐞𝐝 𝐢𝐧 𝐂𝐕𝐏𝐑, 𝐌𝐈𝐂𝐂𝐀𝐈, 𝐖𝐀𝐂𝐕, 𝐓𝐏𝐀𝐌𝐈, 𝐈𝐉𝐂𝐕 & 𝐈𝐂𝐑𝐀 I am a US trained Computer Vision Scientist and Senior AI/ML Engineer with 15+ years of experience building, training, optimizing, and deploying advanced AI systems for real world applications. My core expertise spans state of the art object detection, segmentation, multi object tracking, video understanding, Vision Language Models, Generative AI, and real time video analytics. I go beyond plug and play pretrained models. I solve the hard problems: occlusions, camera movement, crowded scenes, small objects, long duration video, limited data, domain shifts, real time inference, multi camera setups, and production scale deployment. I also build Generative AI and Agentic AI systems that combine computer vision, LLMs, RAG, multimodal reasoning, external tools, and autonomous workflows. 𝐂𝐨𝐦𝐩𝐮𝐭𝐞𝐫 𝐕𝐢𝐬𝐢𝐨𝐧 & 𝐕𝐢𝐝𝐞𝐨 𝐀𝐈 Object detection, classification, segmentation, tracking, open vocabulary detection, action recognition, sports analytics, CCTV/surveillance analytics, crowd and people counting, vehicle and traffic analytics, medical imaging, OCR/document AI, anomaly detection and visual search. 𝐌𝐨𝐝𝐞𝐥𝐬 & 𝐓𝐨𝐨𝐥𝐬 YOLO, RF-DETR, D-FINE, RT-DETR, Grounding DINO, SAM 2/3, Mask R-CNN, DINOv3, CLIP, SigLIP, ByteTrack, BoT-SORT, DeepSORT and custom CNN/transformer architectures. 𝐑𝐞𝐚𝐥 𝐓𝐢𝐦𝐞 𝐕𝐢𝐝𝐞𝐨 & 𝐑𝐓𝐒𝐏 I design scalable pipelines for live cameras, CCTV/DVR/NVR networks and IP cameras using GStreamer, FFmpeg and NVIDIA DeepStream with GPU accelerated inference and edge/cloud deployment. Used across surveillance, retail, sports, manufacturing, transportation and healthcare. 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈, 𝐋𝐋𝐌𝐬 & 𝐀𝐠𝐞𝐧𝐭𝐢𝐜 𝐒𝐲𝐬𝐭𝐞𝐦𝐬 Text to image and diffusion pipelines, vision language models, RAG, semantic search, document intelligence and multi agent systems that combine vision, LLM reasoning and tool use for example: 𝐂𝐚𝐦𝐞𝐫𝐚 → 𝐂𝐨𝐦𝐩𝐮𝐭𝐞𝐫 𝐕𝐢𝐬𝐢𝐨𝐧 → 𝐄𝐯𝐞𝐧𝐭 𝐃𝐞𝐭𝐞𝐜𝐭𝐢𝐨𝐧 → 𝐋𝐋𝐌/𝐕𝐋𝐌 𝐑𝐞𝐚𝐬𝐨𝐧𝐢𝐧𝐠 → 𝐀𝐈 𝐀𝐠𝐞𝐧𝐭 → 𝐁𝐮𝐬𝐢𝐧𝐞𝐬𝐬 𝐀𝐜𝐭𝐢𝐨𝐧. 𝐌𝐞𝐝𝐢𝐜𝐚𝐥 𝐈𝐦𝐚𝐠𝐢𝐧𝐠 & 𝐇𝐞𝐚𝐥𝐭𝐡𝐜𝐚𝐫𝐞 𝐀𝐈 CT, MRI, X-ray & ultrasound analysis, tumor and lesion segmentation, medical image classification, and radiology decision support. 𝐓𝐞𝐜𝐡𝐧𝐢𝐜𝐚𝐥 𝐒𝐭𝐚𝐜𝐤 Programming: Python | C++ | MATLAB AI / Deep Learning: PyTorch | TensorFlow | Keras | Transformers Computer Vision: OpenCV | SAM | Grounding DINO | RF-DETR | YOLO | DINOv3 | CLIP | SigLIP Video: RTSP | FFmpeg | GStreamer | NVIDIA DeepStream | OpenCV Video Medical Imaging: SimpleITK | ITK | Medical Image Processing Pipelines Deployment: Docker | FastAPI | Flask | REST APIs | ONNX | TensorRT | Model Serving Production AI: GPU Optimization | Inference Optimization | Edge AI | Scalable ML Pipelines | Model Testing & Monitoring 𝗥𝗘𝗦𝗘𝗔𝗥𝗖𝗛 & 𝗜𝗡𝗗𝗨𝗦𝗧𝗥𝗬 𝗘𝗫𝗣𝗘𝗥𝗜𝗘𝗡𝗖𝗘 My research is published in 𝐂𝐕𝐏𝐑, 𝐌𝐈𝐂𝐂𝐀𝐈, 𝐖𝐀𝐂𝐕, 𝐈𝐄𝐄𝐄 𝐓𝐏𝐀𝐌𝐈, 𝐈𝐉𝐂𝐕 and 𝐈𝐂𝐑𝐀. I've worked as a Principal Machine Learning Engineer and Senior Computer Vision Scientist, contributing to projects with organizations including IARPA, Facebook/Meta, Google, HEC, NIJ, Xerox (USA) and US DTS. 𝐖𝐡𝐲 𝐜𝐥𝐢𝐞𝐧𝐭𝐬 𝐰𝐨𝐫𝐤 𝐰𝐢𝐭𝐡 𝐦𝐞 15+ years of AI and computer vision experience, PhD level research expertise combined with production engineering, real time and large scale video analytics know how and clear communication with reliable delivery. 𝗛𝗢𝗪 𝗜 𝗖𝗔𝗡 𝗛𝗘𝗟𝗣 Whether you need to build a new AI product, improve an existing Computer Vision pipeline, audit a poorly performing model, process live camera streams, develop a multimodal AI system, or integrate LLM/Agentic AI into your product, I can help from research and architecture through implementation and production deployment. If your project requires more than simply running a pretrained model and needs serious Computer Vision, Generative AI, LLM, Agentic AI or real-time video engineering, let’s discuss the problem and build the right solution.