Overview
The role involves working as an Applied Research Scientist in Descript's Research team, focusing on developing specialized AI models for video and audio editing. The team aims to enhance editing workflows with multimodal understanding, applying cutting-edge research to practical applications.
Key Responsibilities
- Build vision-language systems for Descript's editing features.
- Design benchmarks for evaluating editorial quality, considering cost and latency.
- Develop datasets, including synthetic data generation.
- Train specialized models from scratch or fine-tune existing ones.
- Collaborate with engineering teams to take models from prototype to production.
- Identify future research directions for Descript features.
- Optionally publish research work in academic venues.
Requirements
- Proven ability to design and implement deep learning algorithms, shown through publications, open-source work, or shipped models.
- Strong programming skills, particularly in PyTorch.
- Ability to generate new ideas in machine learning and quickly evaluate them.
- Strong experimental judgment and communication skills.
- A PhD or Master's in deep learning or related field, or equivalent experience.
- At least one notable contribution to top-tier publications or production features involving deep learning.
- For more senior roles, a track record of owning research direction and mentoring other team members is expected.
Benefits
- Base salary range: $197,000–$262,500, plus equity and benefits.
- Generous healthcare package, 401k matching program, catered lunches, and flexible vacation time.
Location
This position can be remote within the United States or based in San Francisco, CA. Remote workers can collaborate in-person at select opportunities throughout the year.
How to Apply
Please apply through Descript's career page.