We believe the most promising path to safe AGI is to automate AI research and code generation to improve models and solve alignment more reliably than humans can alone. Our approach is to combine frontier-scale pre-training, domain-specific reinforcement learning, ultra-long context, and inference-time compute to achieve this goal.
Magic
Software Development
Build aligned and more complete AI to accelerate humanity’s progress on the world’s most important problems
About us
Magic is working on frontier-scale code models to build a coworker, not just a copilot. Come join us: http://magic.dev
- Website
-
https://magic.dev/
External link for Magic
- Industry
- Software Development
- Company size
- 2-10 employees
- Headquarters
- San Francisco
- Type
- Privately Held
- Founded
- 2022
Locations
-
Primary
Get directions
San Francisco, US
Employees at Magic
Updates
-
We believe the most promising path to safe AGI is to automate AI research and code generation to improve models and solve alignment more reliably than humans can alone. Our approach is to combine frontier-scale pre-training, domain-specific reinforcement learning, ultra-long context, and inference-time compute to achieve this goal.
-
We believe the most promising path to safe AGI is to automate AI research and code generation to improve models and solve alignment more reliably than humans can alone. Our approach is to combine frontier-scale pre-training, domain-specific reinforcement learning, ultra-long context, and inference-time compute to achieve this goal.
-
We believe the most promising path to safe AGI is to automate AI research and code generation to improve models and solve alignment more reliably than humans can alone. Our approach is to combine frontier-scale pre-training, domain-specific reinforcement learning, ultra-long context, and inference-time compute to achieve this goal.
-
We believe the most promising path to safe AGI is to automate AI research and code generation to improve models and solve alignment more reliably than humans can alone. Our approach is to combine frontier-scale pre-training, domain-specific reinforcement learning, ultra-long context, and inference-time compute to achieve this goal.
-
We believe the most promising path to safe AGI is to automate AI research and code generation to improve models and solve alignment more reliably than humans can alone. Our approach is to combine frontier-scale pre-training, domain-specific reinforcement learning, ultra-long context, and inference-time compute to achieve this goal.
-
We believe the most promising path to safe AGI is to automate AI research and code generation to improve models and solve alignment more reliably than humans can alone. Our approach is to combine frontier-scale pre-training, domain-specific reinforcement learning, ultra-long context, and inference-time compute to achieve this goal.
-
We believe the most promising path to safe AGI is to automate AI research and code generation to improve models and solve alignment more reliably than humans can alone. Our approach is to combine frontier-scale pre-training, domain-specific reinforcement learning, ultra-long context, and inference-time compute to achieve this goal.
-
We believe the most promising path to safe AGI is to automate AI research and code generation to improve models and solve alignment more reliably than humans can alone. Our approach is to combine frontier-scale pre-training, domain-specific reinforcement learning, ultra-long context, and inference-time compute to achieve this goal.
-
We believe the most promising path to safe AGI is to automate AI research and code generation to improve models and solve alignment more reliably than humans can alone. Our approach is to combine frontier-scale pre-training, domain-specific reinforcement learning, ultra-long context, and inference-time compute to achieve this goal.