Outlier is hiring worldwide — AI Trainers, Coding Experts & Writing EvaluatorsFreelance • Fully remote • $15–$60/hr • Work when you wantBrowse Outlier roles on MMagic.ai →
Outlier is hiring worldwide — AI Trainers, Coding Experts & Writing EvaluatorsFreelance • Fully remote • $15–$60/hr • Work when you wantBrowse Outlier roles on MMagic.ai →
D

AI Engineer — Audio & Video (Remote — Worldwide)

Descript$160k – $230kRemote (Global)Posted 2w ago
PyTorchPythonC++RustCUDAONNXWebAssemblyAudio Signal ProcessingComputer VisionGenerative AI

About the role

As an AI Engineer at Descript, you will build the next generation of creative tools used by millions of podcasters and video creators. You will bridge the gap between cutting-edge research and product, deploying generative models that handle everything from voice cloning to automated video editing. Your work will directly impact our core engine, enabling creators to edit media by simply editing text and pushing the boundaries of what is possible in a browser-based creative suite.

Responsibilities

- Design and implement production-grade AI features like Overdub, Studio Sound, and automated green-screen removal. - Train and fine-tune state-of-the-art models for speech enhancement, voice synthesis, and visual scene understanding. - Optimize inference pipelines to ensure high-performance, low-latency performance within the Descript desktop and web applications. - Collaborate with the Product and UI/UX teams to turn raw model outputs into intuitive, "magical" creative workflows. - Monitor and improve model performance in the wild, addressing issues like audio artifacts or video jitter. - Stay abreast of latest research in LLMs, GANs, and Diffusion models to identify opportunities for new creator-centric features.

Requirements

- 4+ years of experience in software engineering with a focus on machine learning or media processing. - Proven track record of deploying large-scale audio (TTS/STT) or video (diffusion/segmentation) models in production environments. - Deep expertise in PyTorch or JAX, along with strong systems programming skills in C++ or Rust for high-performance inference. - Experience with low-latency real-time processing and optimizing models for edge or resource-constrained environments (CoreML, ONNX, TensorRT). - Strong understanding of DSP fundamentals, including sample rates, codecs, and temporal consistency in video. - Academic background (MS/PhD) or equivalent industry experience in Computer Science, focusing on Generative AI or Signal Processing.

Benefits

Equity, healthcare, parental leave.
Disclaimer: MMagic.ai connects talented people with AI companies around the world. While we work hard to feature quality opportunities, we don't independently verify employers, candidates, salaries, or hiring outcomes. We encourage you to research each opportunity and company before applying or making an offer.

Similar roles