Magic AI
Research Engineer
Magic AI, San Francisco, California, United States, 94199
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 pre-training, domain-specific RL, ultra-long context, and inference-time compute to achieve this goal.
About the role:
As a research engineer, you’ll work on training, evaluating, and serving large AI models and new inference-time compute techniques, build internet-scale datasets, and help prototype new research and product ideas. What you might work on:
Optimize inference throughput for novel model architectures Contribute to frameworks used for our research and production workloads Train trillion-parameter models on large GPU clusters Curate post-training datasets to improve targeted capabilities Build out internet-scale data pipelines and crawlers Design, prototype and optimize new model architectures Contribute to research across long-context, inference-time compute, RL, and more What we’re looking for:
Strong general software engineering skills Thorough knowledge of the deep learning literature Experience with pre- and post-training of LLMs Ability to come up with and evaluate research ideas Experience working with large distributed systems Ability to handle large ETL workloads 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. Our culture:
Integrity.
Words and actions should be aligned Hands-on.
At Magic, everyone is building Teamwork.
We move as one team, not
N
individuals Focus.
Safely deploy AGI. Everything else is noise Quality.
Magic should feel like magic Compensation, benefits and perks (US):
Annual salary range: $100K - $550K Equity is a significant part of total compensation, in addition to salary 401(k) plan with 6% salary matching Generous health, dental and vision insurance for you and your dependents Unlimited paid time off Visa sponsorship and relocation stipend to bring you to SF, if possible A small, fast-paced, highly focused team
#J-18808-Ljbffr
As a research engineer, you’ll work on training, evaluating, and serving large AI models and new inference-time compute techniques, build internet-scale datasets, and help prototype new research and product ideas. What you might work on:
Optimize inference throughput for novel model architectures Contribute to frameworks used for our research and production workloads Train trillion-parameter models on large GPU clusters Curate post-training datasets to improve targeted capabilities Build out internet-scale data pipelines and crawlers Design, prototype and optimize new model architectures Contribute to research across long-context, inference-time compute, RL, and more What we’re looking for:
Strong general software engineering skills Thorough knowledge of the deep learning literature Experience with pre- and post-training of LLMs Ability to come up with and evaluate research ideas Experience working with large distributed systems Ability to handle large ETL workloads 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. Our culture:
Integrity.
Words and actions should be aligned Hands-on.
At Magic, everyone is building Teamwork.
We move as one team, not
N
individuals Focus.
Safely deploy AGI. Everything else is noise Quality.
Magic should feel like magic Compensation, benefits and perks (US):
Annual salary range: $100K - $550K Equity is a significant part of total compensation, in addition to salary 401(k) plan with 6% salary matching Generous health, dental and vision insurance for you and your dependents Unlimited paid time off Visa sponsorship and relocation stipend to bring you to SF, if possible A small, fast-paced, highly focused team
#J-18808-Ljbffr