Member of Technical Staff, Simulated Universes
Engineer building and scaling large-scale simulated environments ('worlds') in which reinforcement-learning 'superlearner' agents emerge and train at an AI research company. Day-to-day: designing, implementing, maintaining and scaling simulation infrastructure with distributed server-client architectures, physics engines, GPU-accelerated simulation (e.g. NVIDIA Warp) and low-level languages like R
- Compensation
- Not disclosed
- Location
- London
- Work setup
- Onsite
- Seniority
- Not specified
- Employment
- Not specified
- Experience
- Not specified
- Category
- Software engineering
- Posted
Skills
Must have
- Industry experience with large-scale simulations, in demanding scientific domains or the games industry
- Experience building systems like huge distributed MMOs relying on world parallelism, large-scale physics engines, or logic preserving consistency despite decentralised computation
- Ability to turn ambitious research visions into reality by designing, building and maintaining codebases that support fast research iteration and efficient execution
- Enjoy working closely with researchers and/or designers, turning their blueprints into worlds agents can explore and learn in
Nice to have
- Experience with physics engines
- Experience with GPU-accelerated physics and simulation frameworks, such as NVIDIA Warp
- Experience with lower level languages such as Rust, C or C++
- Experience implementing distributed server-client architectures
- Experience writing efficient GPU code
Full description
Our mission is to make first contact with superintelligence.
We are creating a superlearner that discovers all knowledge from its own experience, from elementary motor skills through to profound intellectual breakthroughs. This superlearning capability - the ability to endlessly discover knowledge and skills, without relying on human data - will be driven by the world’s most powerful reinforcement learning algorithms.
The superlearner is expected to rediscover and then transcend the greatest inventions in human history, such as language, science, mathematics and technology. If successful, this will represent a scientific breakthrough of comparable magnitude to Darwin: where his law explained all life, our law will explain and build all intelligence.
Role brief
Intelligence cannot emerge in a void. It needs a place to grow and flourish. As a Simulated Universes Engineer, you’ll help us craft, shape and maintain the worlds in which superlearners can emerge and thrive. That means making those worlds accurate, reliable and intricate, and scaling them to unprecedented highs in terms of both parallelism and speed. It’s a challenging quest and one that’s critical to our mission. In this role, you will:
- Build the engineering foundations of our new environments; designing, implementing and maintaining them as they grow
- Partner closely with the team creating the blueprints for the worlds we need to build
- Close the loop between blueprint and build, enabling fast, iterative research cycles
- Scale the environments up, while utilising state-of-the-art hardware
About you
You’re an engineer with a passion for creating artificial worlds, and you know what it takes to make them emerge at scale. You may be a good fit if you:
- Bring industry experience with large-scale simulations, whether in demanding scientific domains or in the games industry
- Know how to build systems like huge distributed MMOs that rely on world parallelism, large-scale physics engines, or logic that preserves consistency despite decentralised computation
- Turn ambitious research visions into reality by designing, building and maintaining codebases that support fast research iteration and efficient execution
- Enjoy working closely with researchers and/or "designers", taking their blueprints and turning them into worlds that agents can explore and learn
On the pure engineering side, you’ll likely have experience with the following: