Hengwen Tong is a director-level engineer based in Beijing with 6 years of focused experience in AI frameworks, compiler and runtime work, currently leading TensorFlow and PyTorch efforts at Cambricon. He brings a strong systems background from prior roles developing BSP, Linux kernel and driver stacks at Xiaomi and high-performance network processor software at FutureWei, coupling low-level performance tuning with ML infrastructure. His career spans carrier-grade and embedded platforms (PowerPC, ARM, MIPS) through to modern deep learning runtimes, giving him a rare full-stack perspective on hardware-accelerated AI. Practical performance engineering is a theme—he has hands-on experience with DPDK, VTune and processor verification that informs efficient model deployment. Colleagues would note he prefers direct technical ownership (and politely declines headhunters, per his profile). Trained as an EE at Beijing Institute of Technology, he blends hardware-aware software design with production ML framework delivery.
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Contributions:8 pushes, 6 branches in 9 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.