AI Computer Vision Intern

**ONLY FULL-TIME CANDIDATES PLEASE**

About the Project:

We are building an AI-powered live streaming product that integrates computer vision, real-time video analytics, and mobile browser compatibility for end users. The focus is on using open-source tools and frameworks, deploying lightweight models like YOLO, and enabling users to access the application seamlessly on mobile devices through their web browsers. This is an excellent opportunity to work on cutting-edge AI, video streaming technologies, and web-based deployment.

Role & responsibilities

1. Data Capture & Annotation:

• Utilize OpenCV to capture and preprocess live video streams, including those initiated via mobile browsers.
• Annotate video datasets for AI model training using tools like LabelImg, Roboflow (free tier), or CVAT.
• Prepare datasets optimized for mobile-friendly streaming and browser-based computer vision tasks.

2. Model Training & Development:

• Train and fine-tune YOLO models (e.g., YOLOv5, YOLOv8) for real-time applications, such as object detection and anomaly detection.
• Optimize models for lightweight deployment, ensuring compatibility with web-based environments and mobile devices.

3. Mobile Browser Access Integration:

• Develop a responsive web application accessible via mobile browsers using HTML5, CSS, and JavaScript.
• Integrate YOLO models for real-time inference in browsers using TensorFlow.js, ONNX.js, or WebAssembly for edge AI processing.
• Optimize the video feed for browser-based streaming using WebRTC, HLS, or RTMP to ensure low latency and high quality.
• Enable features like live object detection directly in mobile browsers, ensuring smooth real-time performance.

4. GPU-Accelerated Model Deployment (Open-Source):

• Deploy YOLO models using open-source GPU frameworks like NVIDIA Triton Inference Server, TensorRT, or ONNX Runtime for scalable processing.
• Use CUDA and cuDNN for inference acceleration on GPU-enabled servers and integrate them with mobile browsers for seamless end-user experiences.
• Configure Docker containers for deploying the backend that serves real-time inference results to mobile clients.

5. Live Streaming & Analytics Integration:

• Build pipelines that capture live video from mobile devices and stream it through browsers to perform real-time computer vision tasks.
• Use FFmpeg or GStreamer for processing and transmitting video streams optimized for web browsers.
• Implement video analytics dashboards that display results such as detected objects, bounding boxes, and insights directly on the mobile browser.

6. Product Launch Preparation:

• Conduct usability testing of the web application on various mobile browsers (e.g., Chrome, Safari, Firefox) to ensure compatibility and performance.
• Assist in creating technical documentation, tutorials, and guides for users to access the application via mobile browsers.

Technical Skills:

• Proficiency in Python, JavaScript, and computer vision libraries like OpenCV.
• Experience with YOLO models (e.g., YOLOv5, YOLOv8) for real-time AI applications.
• Knowledge of browser-based AI integration tools like TensorFlow.js, ONNX.js, or WebAssembly.
• Hands-on experience with streaming technologies (FFmpeg, RTMP, HLS, or WebRTC) and web development frameworks (e.g., React, Vue, or Django).

Application:
If you are passionate about building real-time AI solutions, wed love to hear from you! Please send your resume, portfolio (if available) to ms@emazingideas.com

We are looking for candidates in their final semester ONLY and/or who have recently graduated.

Above all, we are looking for a solution minded individual with hunger for solution.