Principal AI Product Engineer

You will define the long-term technical roadmap for inference, evaluations, and post-training. You will lead architecture initiatives, establish engineering standards, measure cost and performance trade-offs, align technical strategy across functions, mentor senior engineers, and represent the technical approach externally.

Responsibilities

• Define the multi-year technical roadmap for inference, evaluations, and post-training
• Lead large-scale serving, GPU efficiency, evaluation, and RL infrastructure initiatives
• Establish API, benchmarking, evaluation, training stability, and performance-testing standards
• Measure cost, latency, throughput, and model-quality trade-offs
• Identify and resolve long-term technical risks
• Align engineering, research, product, and infrastructure technical strategy
• Mentor Staff and Senior AI Engineers
• Represent the technical approach through open source, publications, talks, and partnerships

Requirements

• 10–15 years of engineering experience
• 4+ years of hands-on LLM inference, GPU performance, evaluation, post-training, or RL experience
• Experience defining multi-year strategy for multi-team AI systems
• Deep knowledge of production LLM inference, GPU performance, evaluations, or post-training and RL infrastructure
• Experience architecting large-scale production inference or training platforms
• Knowledge of CUDA or ROCm, accelerator hardware constraints, and distributed computing
• Experience developing technical leaders

Benefits

• Bonus, equity, and/or commission eligibility
• Medical insurance
• Dental insurance
• Vision insurance
• Flexible paid time off
• Parental leave
• Retirement plan participation