Applied Research Scientist, AI Research
DescriptVideo Editing company
San Francisco, United States$261,625–$299,000Senior
Redpoint Ventures
Andreessen Horowitz
Spark Capital
Atelier Ventures
Todd & Rahul Capital
Data & AINew
About the role
TL;DR
Builds generative media synthesis and multimodal understanding systems for AI-powered editing features.
- •Descript's Research team builds generative video and audio models, and multimodal understanding systems, for distinctive features like Video Regenerate, lipsync, video translation, and zero-shot voice cloning.
- •This role focuses on applied research that ships to production.
- •Key Responsibilities Build generative media synthesis models for video and audio.
- •Develop zero-shot voice and roomtone cloning systems.
- •Create multimodal understanding systems for agentic editing features.
- •Address computer-vision-heavy problems like digital human reconstruction.
- •Identify and pursue new research directions that can become Descript features.
- •Requirements Proven ability to design and implement deep learning algorithms.
- •Strong programming skills in PyTorch and/or TensorFlow.
- •Track record of generating new ideas in machine learning.
- •Strong experimental judgment and ability to test ideas quickly.
- •PhD or Master's in deep learning or related field, or equivalent experience.
Required skills
PythonPyTorchTensorFlowComputer VisionNLPLLMs
Domain expertise
aiproductivity
Benefits & perks
Equity, Healthcare package, 401k matching, Catered lunches, Flexible vacation time
Tech stack
PythonPyTorchTensorFlowComputer VisionNLPLLMs