You are a software engineer with a real passion for software creation and problem solving with an enthusiastic, can-do attitude to overcoming technical challenges.
To succeed in this role you need a strong foundation in software engineering and enjoy working on a wide range of diverse and challenging problems within a mission-driven team.
You will be part of the Worlds team building games for AI agents.
DeepMind's mission is to "Solve intelligence. Use it to make the world a better place." The Worlds team provides DeepMind Researchers with the best training and testing environments possible. These range from bespoke mini-games aimed at answering specific research questions, to expansive first-person environments using modern game engines. The team is important in helping steer DeepMind’s research forward.
We design and implement our learning environments carefully as it’s likely we’ll live with them for many years to come so we apply thoughtfulness to everything we do. We take robust design and implementation seriously to ensure the delivery of high quality, repeatable, research environments. You’ll join a close-knit team of accomplished individuals who openly share ideas with one another. Projects we’ve worked on include AlphaGo, DM Lab, and DQN on Atari as well as contributions to countless publications. No machine learning experience is required for this role.
DeepMind's AI systems are the best in the world. They have surprised the world by learning to play video games, mastering the ancient Chinese game of Go, and making inroads into solving protein folding. They are trained to perform visual processing tasks, exhibit cognitive abilities like memory and planning, and can learn to control physical bodies. Much of what you know is of huge value to us as we strive towards building the challenges which will be used as training environments for increasingly general-purpose learning systems, but there will also be much you will need to learn about the process of learning itself.
If this challenge intrigues you, then get in touch!
About The Role
You’ll develop and support our underlying infrastructure to accelerate games creation whilst helping to ensure optimum performance. All games are wrapped with a common environment API so agents can switch between them without interface changes. For Research to proceed as quickly as possible our training environments need to run much faster than real time, at massive scale and across global infrastructure. You will work collaboratively within the team and beyond to help build, maintain and optimise our environments to ensure that they do.
We’re looking both for engineers who are passionate about people management, and those who want to stay purely technical. We’re careful to balance responsibilities such that all engineers on the team, whether managers or not, can contribute directly.
- Strong software engineer with extensive experience in software design
- Proven C++ or C# enterprise programming experience; professional or otherwise
- Experience with concurrent or distributed software architecture
- Python experience ideal, but not essential
- Rendering experience useful
- Experience with gameplay, engine, graphics, build systems or physics programming
- Basic understanding of maths and statistics
To accompany your application, please submit a covering letter, answering the following questions:
- Why are you interested in DeepMind?
- What interests you about the projects DeepMind has worked on
- What game of your own are you most proud of and why?