Gemini Post-Training ML Engineer — RL & Multimodal AI

Google Inc.; New York (NY); Hybrid; USD 174,000 - 252,000; Full time

Job summary

Google DeepMind seeks a Gemini Post-training Software Engineer to advance AI agents and multi-modal systems. You will design, train, and evaluate models for production readiness across Gemini projects, with emphasis on post-training tuning and deployment.

This role requires a CS degree, 5 years ML experience, Python, ML testing, and background with NLP/Generative AI and LLMs. You will join a world-class AI lab focused on responsible innovation and scalable systems.

Qualifications

• Bachelor’s degree in Computer Science, a similar field, or equivalent practical experience.
• 5 years of experience designing, training, and evaluating machine learning models.
• Experience in Python programming.
• Experience in test writing for ML purposes (TensorFlow or similar).
• Experience with Large Language Models, NLP, or Generative AI.

Responsibilities

• Make core contributions to Gemini that advance AI agents.
• Identify model areas of improvement.
• Train and evaluate agents across digital environments.
• Focus on multi-step reinforcement learning in multimodal environments.

Skills

Python

ML models

ML testing

NLP

Generative AI

Education

Bachelor's degree in Computer Science

Tools

TensorFlow

Job description

Google DeepMind seeks a Gemini Post-training Software Engineer to advance AI agents and multi-modal systems. You will design, train, and evaluate models for production readiness across Gemini projects, with emphasis on post-training tuning and deployment.

This role requires a CS degree, 5 years ML experience, Python, ML testing, and background with NLP/Generative AI and LLMs. You will join a world-class AI lab focused on responsible innovation and scalable systems.