Job Title: Senior VR/XR & Computer Vision Software Engineer Company Name: IVN Inc. Job Url: https://www.simplyhired.com/job/yYJktacJss_ns4_bRQpZEzJzG76yhh1F99uTM4VMsCooAvicXeZw0w Job Description: Senior VR/XR & Computer Vision Software Engineer IVN Inc. Remote Job Details Contract $30 - $40 an hour 13 hours ago Qualifications GLSL TensorFlow DevSecOps Practices Performance tuning Node.js Amazon EC2 PyTorch OAuth 2.0 Computer vision Kanban C# 5G Scalable systems Windows Infrastructure architecture Java Revit C++ Solution architecture design Docker SSL Model deployment PostgreSQL Terraform Redis Arrays (data structure) Prototype creation Ubuntu Django Cloud Native Design OpenCV Scalability Developing and maintaining backend systems Regression analysis WebRTC Agile software development S3 Vue.js Machine learning libraries Flask Real-time data processing implementation Senior level Rendering AI Cross-functional collaboration Cross-validation Machine learning frameworks TypeScript Prototypes NGINX Generative AI Debugging Cross-functional communication MySQL Route 53 AWS CloudFormation Full Job Description Job Posting: Senior VR/XR & Computer Vision Software Engineer With Experience Building Full-Stack AI-Driven Immersive Systems Location: Remote (AWS-Cloud Focused) Infrastructure: Ubuntu 24.04, AWS (G5/C6g), Red5 Pro (Streaming Server, TrueTime, and Terraform), NGINX Stack: TensorFlow, Three.js / Babylon.js, React / Vue.js / Django, Node.js What You’ll Build & Prototype AI‑Driven Computer Vision & Facial Tracking Rapidly prototype and refine facial landmark detection pipelines using MediaPipe, ARKit/ARCore, or custom TensorFlow/PyTorch models. Build ML‑based blendshape mapping systems (regression, PCA, neural networks) based on live-streaming data that drive expressive 2D and 3D full-body representation of a person. Experiment with new Cross-Validation/Machine Learning techniques and integrate them into real‑time pipelines with sub‑100‑ms latency. Use AI tools to accelerate model iteration, data labeling, and performance tuning. AI Integration, implementation, and development of fine-tune tracking models using Google MediaPipe, TensorFlow, ARKit, and NVIDIA Maxine. 3D Graphics, Avatar Animation & Blender Integration Build expressive 3D environments and animation systems using Three.js. Work directly with Blender to create, modify, and export glTF avatars with morph targets and PBR materials. Prototype new animation techniques, shader effects, and environment interactions quickly. Shader Programming by writing custom GLSL/HLSL shaders to handle realistic skin textures, PBR materials, and real-time lighting within the virtual space. 3D Environment Development: Architect and render realistic 3D worlds using Three.js, Babylon.js, and Blender ensuring high-performance execution via WebGL and WebGPU. Collaborate with artists and designers in a fast feedback loop. Real‑Time WebRTC & Distributed Media Systems Architect WebRTC pipelines for video, audio, and data channels. Integrate with our AWS‑hosted Red5 Pro cluster, including Streaming Manager, TrueTime Meeting, Terraform, and autoscaling components. Prototype new networking strategies, optimize ICE/TURN/STUN flows, and experiment with adaptive bitrate logic. Use AI‑assisted debugging and monitoring tools to accelerate iteration. Develop live real-time avatars and optimize the pipeline for mapping 2D webcam/camera feeds to 3D meshes using Facial Action Coding System (FACS) and morph target influences. Startup‑Style Collaboration Work closely with founders and cross‑functional teammates to shape product direction. Prototype features rapidly, test them with real users, and iterate based on feedback. Contribute to architectural decisions, technical strategy, and long‑term platform vision. Thrive in a culture that values creativity, autonomy, and speed. Role Overview We are seeking a high-caliber Senior Full-Stack Software Engineer/Developer with a specialization in VR/XR and Computer Vision to lead the engineering of our "metaverse-class" social immersion platform. This role is at the bleeding edge of technology: you will build the systems that overlay live-streamed facial video onto computer-generated and live-captured 2D and 3D images of a person, creating a seamless "hybrid" user representation in realistic 3D worlds. As a Full-Stack lead, you will own the entire vertical—from the Ubuntu/AWS backend and Red5 Pro streaming pipelines to the TensorFlow tracking models and the React/Three.js frontend environments. 1. Full-Stack Responsibilities & Frameworks · Frontend (Web & VR/XR): Develop high-performance, immersive interfaces using React or Vue.js. Implement the 3D world layer using Three.js or Babylon.js, ensuring the hybrid-user representation renders with sub-100ms latency via WebGL. · Backend & API: Architect robust server-side logic using Django or Node.js to manage user sessions, world state synchronization, and secure data persistence, Red5 Pro streaming server, NGINX (reverse proxy, SSL, caching), Docker/Kubernetes (scaling), and Redis/PostgreSQL for session/state management. · Streaming Infrastructure: Manage the Red5 Pro TrueTime Meeting and Stream Manager deployment on Ubuntu 24.04, handling the ingestion and distribution of ultra-low latency WebRTC streams. · Native Integration: Maintain and extend client functionality for Windows and Mobile using C# or C++. 2. AI, Computer Vision & Prototyping · CV/ML Pipeline: Develop and integrate TensorFlow (and TensorFlow.js) models for real-time facial landmarking. Use Cross-Validation to ensure tracking accuracy across all user demographics. · AI-Expedited Dev: Utilize AI tools (GitHub Copilot, Gemini, Cursor, v0.dev, etc.) to accelerate full-stack feature delivery and Luma AI for rapid 3D environment asset generation. · Hybrid Rendering: Write custom GLSL/HLSL shaders to blend live video textures onto 2D and 3D meshes realistically. 3. Methodology & Security (Secure SDLC) · Secure SDLC: Apply a "Security by Design" philosophy across the full stack or critical components of the stack. Perform threat modeling on WebRTC signaling, implement secure OAuth2/OIDC flows, and manage NGINX for SSL termination and WAF protection. · Agile Methodology: Lead and participate in Agile workflows (Scrum/Kanban), translating complex user stories into technical requirements and iterative releases. · DevOps: Orchestrate AWS infrastructure (EC2, S3, CloudFormation) and optimize the Ubuntu server environment for high-throughput media processing. 4. Technical Toolbox Category Requirements Frontend Frameworks React, Vue.js, Angular, Three.js, Babylon.js Backend Frameworks Django, Node.js, FastAPI Streaming & Network Red5 Pro (TrueTime & Streaming Manager API), WebRTC, RTSP, NGINX, Ubuntu 24.04 AI & CV TensorFlow, OpenCV, MediaPipe, Scikit-learn Cloud (AWS) EC2 (G5/C6g), S3, Route53, AWS Wavelength (5G Edge) Languages Python, JavaScript/TypeScript, Java, C#, C++ Methodologies Agile, Secure SSDLC DevOps AWS CloudFormation/Terraform 3D/Graphics Three.js, GLSL Shaders, UV Mapping, Blender Databases Mongo, MySQL, NoSQL, Redis/PostgreSQL Key Qualifications · Experience: 7+ years in Full-Stack development, with at least 3 years focused on Real-Time Communications (RTC), VR/XR, WebRTC, and Computer Vision. · Architecture: Proven ability to design and deploy scalable, secure client-server cloud-native applications on AWS, VPS, and bare-metal servers. · Problem Solving: Deep understanding of the synchronization challenges between live video streams and 3D coordinate data. · Security: Solid experience implementing security best practices throughout the software development life cycle. · Portfolio demonstrating AI-driven video manipulation or Web3D immersive environments. How to Apply Submit your resume to support@ivn.net with a link to your GitHub or portfolio. We are specifically looking for candidates who can demonstrate a completed project involving real-time video integration using Red5 Pro implementations on AWS platform, TensorFlow-based tracking, 3D web rendering, and a secure backend architecture. Pay: $30.00 - $40.00 per hour Work Location: Remote