Principal M-LLM Post-Training & Inference Engineer

engineeringjobs.net, Inc.; Town of Montana (WI); On-site; USD 307,000 - 427,000; Full time

Job summary

Google is hiring a Principal SWE for Multimodal LLM post-training and execution in the XR space. You will lead production‑ready features, ensure safety, and optimize real-time XR performance while pushing the boundaries of multimodal AI in immersive environments.

The role sits at the intersection of model alignment and high‑performance execution, requiring deep expertise in large‑scale ML systems and a track record of leading engineering teams through full lifecycle delivery.

Qualifications

• Bachelor's degree in CS, Math, or equivalent practical experience.
• 15 years of software engineering experience in large-scale ML systems.
• Experience driving go-to-market strategy for AI technologies.
• Experience leading engineering teams through full AI product execution.

Responsibilities

• Design and scale post-training infrastructure for Multimodal LLMs, including instruction tuning, RLHF, and evaluation frameworks.
• Architect and optimize execution engines for real-time M-LLM inference on XR hardware.
• Lead data pipelines for computer vision and multimodal understanding of spatial contexts.
• Collaborate with Product and UX to define intelligence primitives for spatial XR applications.
• Set safety and alignment standards for XR AI deployment at scale.

Skills

Software engineering

Leadership

Go-to-market strategy

Education

Bachelor's degree

Tools

RLHF workflows

AI accelerators

Job description

Google is hiring a Principal SWE for Multimodal LLM post-training and execution in the XR space. You will lead production‑ready features, ensure safety, and optimize real-time XR performance while pushing the boundaries of multimodal AI in immersive environments.

The role sits at the intersection of model alignment and high‑performance execution, requiring deep expertise in large‑scale ML systems and a track record of leading engineering teams through full lifecycle delivery.