Teams at Sciforium

Design, optimize, and integrate custom GPU kernels for large-scale AI training and inference. The role involves low-level C++ and PTX development, CUDA/ROCm optimization, framework integration with PyTorch and JAX, performance profiling, bottleneck analysis, benchmarking, and collaboration with ML, distributed systems, serving, and hardware teams.

The GPU Kernel Engineer will design and optimize GPU kernels for large-scale AI systems, integrating them into ML frameworks and collaborating across teams to enhance performance.

Develop and pre-train large byte-native and multimodal foundation models. Implement architectures, training objectives, optimization methods, and stable scaling recipes. Build production-grade training infrastructure, run ablations, analyze training dynamics, and improve model quality, efficiency, reliability, and scalability across GPU-based distributed environments.