Member of Technical Staff, Compute
About The Position
Mirendil is a tech-first company focused on solving core bottlenecks that unlock step-change acceleration across science and technology. Its first goal is to democratize frontier AI research and development across scientific disciplines. The company is building a frontier AI research company and training its own models end-to-end. The Compute role focuses on making compute capacity available and reliable for research. The engineer will integrate GPUs across multiple providers into a unified fleet that researchers can use without managing the underlying infrastructure. The work will affect how much compute is actually used for research and may include bringing up and maintaining clusters, scheduling and allocating resources across pretraining, reinforcement learning, and inference, improving reliability, planning future capacity with research teams, and increasing GPU utilization.
Responsibilities
- Bring new clusters from delivery to running jobs quickly and maintain cluster health over time.
- Schedule and allocate compute so pretraining, reinforcement learning, and inference can share capacity effectively.
- Detect failing hardware and route around it automatically to reduce the risk of losing training runs.
- Plan compute needs with research teams and help decide what compute to buy, from whom, and when.
- Measure the research output of GPU-hours and improve utilization.
Benefits
- Competitive benefits
- Meaningful equity grant
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Mid Level
Education Level
No Education Listed
