ML Performance & Kernel Engineer

ByteDance is seeking a talented engineer to optimize the computational performance of its recommendation mid-platform models. You will design high-performance kernel libraries and drive compilation enhancements, collaborating with multiple teams to diagnose bottlenecks and improve efficiency across training and inference. The role requires strong C/C++ skills, knowledge of CPU/GPU architectures, and hands-on experience with model graph optimization and compiler stacks like TVM/MLIR/XLA.#J-18808-Ljbffr