Software Engineer, AI Inference / HPC

Job description

54,000 new photos are taken every second, and 600 hours of video are uploaded every minute. At Topaz Labs, we help over 1 million paying customers (including teams at Google, Nvidia, and NASA) maximize the visual quality of over 1 billion of these photos and videos. **About us** Topaz Labs is a full-stack AI company that develops, trains, and deploys generative AI models for image and video enhancement. We’re the best in the world at improving image and video quality, and produce several award-winning desktop products that millions of people rely on. **About the role** As a Software Engineer supporting our AI Engine, you would report to our Head of AI Engine, and use your expertise to improve performance of our internal AI Engine. Your responsibilities will be to increase app performance, stability, availability of new features and simplify and improve the API of the framework. You will be the technical bridge between our Deep Learning research team and our variety of Production products that rely on our models. You’ll support our research team by helping prepare new & updated models for production and helping with GPU/CPU optimization. You’ll also work with various hardware partners (NVIDIA, AMD, Intel, Apple) to optimize inference on their hardware. ### About you - Hands-on experience with performance optimization, e.g. concurrency, multithreading, memory, speed, benchmarking, reliability - Experience architecting APIs for internal development - Hands-on experience implementing image processing or computational photography algorithms - Expert knowledge of C/C++ - At least 1+ years of professional working experience in a related field ### Preferred - Experience with video encoding/decoding and file formats - Experience with OpenCV, ffmpeg, GPU programming - Experience with the raw image camera pipeline and image formats - Experience with onnx, coreml, and tensorRT runtime SDKs - Interest in photography or videography