We’re seeking someone to build out a new team - in this role you’ll be responsible for growing and leading a hybrid research-engineering group that jointly optimizes for availability, latency, and throughput of our model inference service. You would be the first official hire into this team — you would help to grow it by hiring and training new members.

The ideal candidate is someone who has built and scaled production services, has strong intuitions for performance engineering, and has successfully led large engineering projects in the past - ideally with a research flavor. This service is completely critical to our product - reductions in latency can result in nonlinear increases in user satisfaction, so the role has tremendous leverage overall.

Responsibilities:

• Build and manage a hybrid team of researchers and engineers
• Build, operate, and maintain the core inference service, optimizing for uptime, latency, and throughput
• Collaborate with other research teams propose and implement research ideas that translate into latency improvements
• Build software to safely and seamlessly translate trained model checkpoints into efficiently serve-able artifacts

Qualifications:

• Strong leadership and people management skills, with a track record of leading successful research or engineering projects to completion
• Intuition for performance engineering - latency really matters for our use-case
• Excellent communication and collaboration skills, with the ability to work effectively with both other research stakeholders and the core infrastructure team
• Experience building and scaling production services, ideally in a high-growth environment
• Experience with hardware accelerators in data centers

The pay range for this position in California is $175,000 - $350,000yr; however, base pay offered may vary depending on job-related knowledge, skills, candidate location, and experience. In addition to base salary, we also offer competitive equity and benefits packages.