Descript is powered by state-of-the-art Deep Learning technology, lowering the barrier to entry for content creation. Our vision is to build the next generation platform to enable easy and fast creation of audio and video content powered by cutting-edge AI. Building a revolutionary way to record, transcribe, edit and mix spoken audio and video comes with a series of unique challenges and requires solving hard and complex problems. This team has pioneered text-based media editing through tools like Overdub and Studio Sound. 

As a member of this team you will:

• Own the ML infrastructure and production pipelines to bring the magic of AI to the end users
• Collaborate with the engineering team to bring  AI services to life
• Provide guidance on the performance trade-offs of Deep Learning models in production to the rest of the organization

Requirements

• Ability to write flawless, readable and maintainable code in Python. Experience with Go is a plus.
• Familiarity with gRPC and Deep Learning model deployments
• Experience in design, development and maintenance of tools for ML data and modeling pipelines, including big data processing, model training and evaluation, continuous integration and development, model quality monitoring and analytics
• Previous experience with developing machine learning infrastructure or experience in using GPUs
• Strong analytical skills and a real passion for efficiency of the data and modeling process
• Familiarity with deep learning frameworks (e.g. Tensorflow, PyTorch)
• Preferred: Experience with on-device deployment of the deep learning models
• Preferred: Ability to deal well with ambiguous problems, and proceed with good judgement
• Preferred: Knowledge and experience with Google Cloud Platform and Kubernetes