This role focuses on large-scale model training across massive GPU clusters. You will work at the boundary between deep learning and distributed systems, ensuring that training runs are performant, reliable, and reproducible under extreme scale.
Magic’s long-context models create non-trivial systems challenges: sustained memory pressure, communication overhead across thousands of devices, long-running jobs that must survive failures, and efficient sequence packing under hardware constraints. You will own the systems that make large-scale pretraining stable and fast.
Magic’s mission is to build safe AGI that accelerates humanity’s progress on the world’s most important problems. We believe the most promising path to safe AGI lies in automating research and code generation to improve models and solve alignment more reliably than humans can alone. Our approach combines frontier-scale pretraining, domain-specific RL, ultra-long context, and inference-time compute to achieve this goal.
Magic strives to be the place where high-potential individuals can do their best work. We value quick learning and grit just as much as skill and experience.
ReflectionSan Francisco, CA +2
Pay not posted
Posted Mar 24, 2026
$150K–$450K / yr
Posted Oct 5, 2026
Engram LabSan Francisco
Pay not posted
Posted Jun 23, 2026
$400K–$500K / yr
Posted Aug 4, 2026
Pay not posted
Posted Aug 1, 2026
MindbeamUnited States
$150K–$190K / yr
Posted Jul 4, 2026
Source: the Magic Ashby job board. We last checked this job 2 h ago. We first saw it 8 days ago. A job on a job board does not always mean that the team interviews people now. We found this job board in a public list. We did not yet compare it with the careers page of the company.
No one has reported an interview at Magic yet. Did you interview there? Add the questions.
This base pay is above the median for inference and ML systems jobs in the Bay Area with no level in the job title.
We compare the middle of each posted range. This is base pay as posted. The group does not include this job.