We’re looking for a talented Compiler Engineer to help accelerate world-class AI research and make a difference! The work that we do presents interesting and unique engineering challenges and as a member of the DeepMind's engineering team, you’ll contribute towards enhancing the performance of our research infrastructure and learning algorithms.
These machine learning algorithms are becoming increasingly complex, with computational requirements growing at a rapid pace. Often, these demands can only be met by hardware accelerators, such as GPUs and TPUs. Utilising these processors efficiently is challenging and algorithms are constantly evolving, making hand-crafted optimisation infeasible in many cases.
This is why we are looking for people that can help us work on more general solutions that can adapt to evolving hardware and software. If you’re a compiler engineer then this is a career-changing opportunity! You'll collaborate with engineers at DeepMind and across Google as well as contribute to compiler technologies. Importantly, you'll use your expertise to identify and develop techniques to map new algorithms to hardware accelerators.
- Good communication skills
- Experience with LLVM
- Experience with C++
- Experience with parallel programming
- Experience with performance profiling
- Knowledge of CUDA or OpenCL
- Experience with GPUs or other hardware accelerators
- Knowledge of Python
- Interest in AI