🧠 What You’ll Build

A privacy-first, real-time mobile SDK for facial analysis and avatar control

Lightweight neural inference pipeline optimized for CoreML (iOS) and NNAPI/TFLite (Android)

Media pipeline integrations with camera, mic, video overlays, and audio triggers

Expression and mood detection APIs for apps to react to user context in real-time

Interfaces to Unity/WebRTC/OpenGL for twin rendering and livestream overlays

🛠️ Responsibilities

Architect and implement performant on-device SDKs in Swift/Kotlin/C++

Integrate model runtimes (e.g. CoreML, TensorFlow Lite, MNN, MediaPipe)

Build developer-facing APIs and documentation for 3rd-party integration

Profile, test, and optimize for battery, latency, and device diversity

Work with backend and avatar teams to ensure seamless client–cloud coordination

Collaborate with product and design on UX for facial feedback and avatar mirroring

🔍 Ideal Experience

4+ years in mobile SDK or media app development (iOS and/or Android)

Deep experience in camera APIs, video/audio processing, and real-time ML inference

Familiarity with WebRTC, OpenCV, Metal/Vulkan, Unity integration

Exposure to facial landmarking, pose estimation, or emotion classification models

Prior work on communication apps, AR filters, or avatar systems is a big plus

🚀 Bonus Points

Experience with custom ML models (ONNX, MNN, PyTorch Mobile)

Understanding of privacy-first design principles and edge-device ML constraints

Prior contribution to a widely used mobile SDK

🌐 Technologies You’ll Use

Swift / Objective-C / Kotlin / Java / C++

CoreML / TFLite / ONNX / MediaPipe / ARKit / MLKit

• WebRTC, OpenGL, Metal, AVFoundation, CameraX