Research Engineer: Agentic AI & Frontier Models

Socket.dev; Mountain View (CA); On-site; USD 174,000 - 252,000; Full time

Job summary

Google is seeking a research-focused Software Engineer in Mountain View, CA to drive agentic AI initiatives and scale experiments across Gemini frontier research. You will prototype architectures, collaborate with universities, and contribute to research publications while delivering tangible AI capabilities.

Responsibilities include designing agent environments, developing evaluators, and turning trajectories into high-quality data for SFT and RL post-training to advance Google’s AI roadmap.

Qualifications

• 2 years of experience with agentic AI workflows, frameworks, and approaches.
• 1 year of experience with machine learning and deep learning.
• Experience in software engineering and cloud-based development.
• Experience with Python, TensorFlow, PyTorch, or similar ML frameworks.

Responsibilities

• Design, build, and scale agent environments and task suites; research and develop Agentic AutoRaters and benchmark against frontier models.
• Enable agents to automatically generate verifiers and iterate on them for effective exploration.
• Develop trajectory analysis frameworks and diagnostic tooling to identify root-cause agent failure modes and optimize harnesses.
• Harvest complex multi-turn trajectories into datasets and reward signals to power SFT and RL flywheels for Gemini models.

Skills

Agentic AI

ML experience

Software engineering

Python & ML frameworks

Tools

Python

TensorFlow

PyTorch

Job description

Google is seeking a research-focused Software Engineer in Mountain View, CA to drive agentic AI initiatives and scale experiments across Gemini frontier research. You will prototype architectures, collaborate with universities, and contribute to research publications while delivering tangible AI capabilities.

Responsibilities include designing agent environments, developing evaluators, and turning trajectories into high-quality data for SFT and RL post-training to advance Google’s AI roadmap.