Build the core infrastructure to serve thousands, then millions, of private, personalized AI models at scale for an early-stage AI platform, ensuring user data and models remain confidential.
Requirements
- Deep understanding of the machine learning stack, including transformer optimization and GPU performance.
- Ability to execute quickly in a fast-paced, early-stage startup environment.
Responsibilities
- Build infrastructure for deploying thousands to millions of personalized finetuned models.
- Monitor and optimize in-the-wild model serving performance for low latency and cost.
- Integrate with a TEE-based privacy stack to guarantee user data and model confidentiality.
- Optimize model serving performance for low latency and cost.
- Drive the foundational systems for a new era of personal AI.
Other
- A missionary mentality, passionate about ensuring AI works for people.
