Member of Technical Staff (GPU Performance Engineer)

Stack

Responsibilities

• ▹Design and implement improvements to the training infrastructure
• ▹Directly contribute to technical decisions that optimize model performance
• ▹Work on post-training processes, including reinforcement learning and fine-tuning
• ▹Improve the efficiency and scalability of model serving infrastructure

Requirements

• ▹Strong engineering skills with fluency in Python and PyTorch (or other frameworks)
• ▹Proven experience implementing and training large deep learning models
• ▹Experience writing and debugging low-level GPU code (CUDA, C++)
• ▹Experience scaling GPU jobs on large-scale compute clusters (e.g., Slurm or Kubernetes)
• ▹Demonstrated ability to analyze and optimize performance of GPU-accelerated workloads (profiling, bottleneck identification, performance tuning)

Soft skills

What we offer

• ▹Elite team: collaborate with engineers and researchers from Google DeepMind, Facebook AI Research (FAIR), and successful startups
• ▹Cutting-edge infrastructure: train state-of-the-art models on the latest software and hardware
• ▹Inclusive and open culture
• ▹5 weeks of paid leave
• ▹Comprehensive healthcare including vision and dental
• ▹Visa support (H1B, OPT transfer) for US employees

About the company

Reka's mission is to build useful multimodal artificial intelligence to empower organizations and businesses. A globally distributed, remote-first foundation model startup headquartered in the San Francisco Bay Area, its team includes talent from around the world, including alumni of Google DeepMind and Facebook AI Research (FAIR).