Warsaw Engine Team is responsible for making speech-related models achieving state-of-the-art performance. It is a mixture of solid development, optimisation and R&D work. We are looking for candidates to join our team who will:
- Optimise deep learning-related code for all common platforms, including CUDA, x86, ARM and others
- Develop and maintain our in-house deep learning compiler
- Develop and maintain our in-house inference engine
- R&D over achieving state-of-the-art performance.
- Ability to carry on development effort independently
- Proficient in C/C++
- Proven experience in optimising any software for any hardware
- At least basic ML/DL knowledge
- Computer architecture knowledge
- Experience with any SIMD development (SSE, AVX, NEON) is a plus
- Experience with GPU development (CUDA, OpenCL, Vulkan, OpenGL, Direct3D) is a plus
- Experience with compiler-related development (implementing DSLs, JIT assemblers, experience with LLVM or DL compilers like TVM or Halide) is a plus.