What you'd actually do

Architect and implement advanced Reinforcement Learning (RL) workflows for complex, multi-turn agentic tasks. Develop novel training recipes for reasoning, self-correction, and tool use (e.g., CoT, Tree of Thoughts) to improve model reliability in long-horizon workflows.

Design robust reward systems and simulation environments ("Digital Twins") to evaluate and train agents.

Create the "Intelligence Assets" required to train specialized student models, bridging the gap between generalist teacher models and domain-specific production requirements.

Contribute to the unified middleware layer that democratizes access to state-of-the-art tuning. Implement efficient adaptation techniques (e.g., LoRA, Distillation, Quantization) to ensure high-performance agents can be deployed under strict latency and cost constraints.

Partner with Google DeepMind researchers to validate novel algorithmic approaches (e.g., outcome-supervised vs. process-supervised RMs) and scale them from 0-to-1 prototypes into 1-to-N production libraries used across Google.

Skills

Required

Python
JAX
PyTorch
Reinforcement Learning (RLHF, RLAIF)
LLM post-training techniques (SFT, DPO, PPO)
Large Language Models (LLMs)
Multi-Modal
Large Vision Models
model deployment
model evaluation
data processing
debugging
fine tuning

Nice to have

Master’s degree or PhD in Engineering, Computer Science, or a related technical field
8 years of experience with data structures and algorithms
3 years of experience in a technical leadership role leading project teams and setting technical direction
multimodal learning
embodied agents
reward modeling
dense/mixture of experts (MoE) architectures
hybrid reward systems

What the JD emphasized

8 years of experience in software development

5 years of experience leading ML design and optimizing ML infrastructure

2 years of experience with GenAI techniques

Experience in Reinforcement Learning (RLHF, RLAIF) and LLM post-training techniques (SFT, DPO, PPO)

Experience building efficient evaluation harnesses, benchmarks, or simulation environments for measuring agent performance

Proven track record (publications or production launches) in reward modeling

Other signals

architecting the technical bridge between frontier research and massive-scale product deployment

pioneer the next generation of Agentic Reinforcement Learning

architect the "cognitive" layer of Google’s AI stack

translate frontier research into scalable production infrastructure

solving the "GenAI Engineering Gap" by transforming probabilistic models into reliable, self-improving agentic systems

Google's software engineers develop the next-generation technologies that change how billions of users connect, explore, and interact with information and one another. Our products need to handle information at massive scale, and extend well beyond web search. We're looking for engineers who bring fresh ideas from all areas, including information retrieval, distributed computing, large-scale system design, networking and data storage, security, artificial intelligence, natural language processing, UI design and mobile; the list goes on and is growing every day. As a software engineer, you will work on a specific project critical to Google’s needs with opportunities to switch teams and projects as you and our fast-paced business grow and evolve. We need our engineers to be versatile, display leadership qualities and be enthusiastic to take on new problems across the full-stack as we continue to push technology forward.

With your technical expertise you will manage project priorities, deadlines, and deliverables. You will design, develop, test, deploy, maintain, and enhance software solutions.

Domain Applied ML (DAML) operates as Google’s "Applied AI Layer," architecting the technical bridge between Google DeepMind’s frontier research and massive-scale product deployment. We define the company-wide strategy for Foundation Model adoption and engineer high-performance solutions in critical domains.

In this role, you will pioneer the next generation of Agentic Reinforcement Learning. You will architect the "cognitive" layer of Google’s AI stack—developing novel RL recipes, reward modeling systems, and synthetic data flywheels that enable models to reason, plan, and use tools effectively. You will translate frontier research into scalable production infrastructure, solving the "GenAI Engineering Gap" by transforming probabilistic models into reliable, self-improving agentic systems.

Google Cloud accelerates every organization’s ability to digitally transform its business and industry. We deliver enterprise-grade solutions that leverage Google’s technology, and tools that help developers build more sustainably. Customers in more than 200 countries and territories turn to Google Cloud as their trusted partner to enable growth and solve their most critical business problems.

The US base salary range for this full-time position is $207,000-$300,000 + bonus + equity + benefits. Our salary ranges are determined by role, level, and location. Within the range, individual pay is determined by work location and additional factors, including job-related skills, experience, and relevant education or training. Your recruiter can share more about the specific salary range for your preferred location during the hiring process.

Please note that the compensation details listed in US role postings reflect the base salary only, and do not include bonus, equity, or benefits. Learn more about benefits at Google.

Responsibilities

Architect and implement advanced Reinforcement Learning (RL) workflows for complex, multi-turn agentic tasks. Develop novel training recipes for reasoning, self-correction, and tool use (e.g., CoT, Tree of Thoughts) to improve model reliability in long-horizon workflows.
Design robust reward systems and simulation environments ("Digital Twins") to evaluate and train agents.
Create the "Intelligence Assets" required to train specialized student models, bridging the gap between generalist teacher models and domain-specific production requirements.
Contribute to the unified middleware layer that democratizes access to state-of-the-art tuning. Implement efficient adaptation techniques (e.g., LoRA, Distillation, Quantization) to ensure high-performance agents can be deployed under strict latency and cost constraints.
Partner with Google DeepMind researchers to validate novel algorithmic approaches (e.g., outcome-supervised vs. process-supervised RMs) and scale them from 0-to-1 prototypes into 1-to-N production libraries used across Google.

Qualifications

Minimum qualifications:

Bachelor's degree or equivalent practical experience.
8 years of experience in software development.
5 years of experience leading ML design and optimizing ML infrastructure (e.g., model deployment, model evaluation, data processing, debugging, fine tuning).
2 years of experience with GenAI techniques (e.g., Large Language Models (LLMs), Multi-Modal, Large Vision Models) or with GenAI-related concepts (language modeling, computer vision).
Experience in Python and with ML frameworks (JAX, PyTorch) for large-scale model training.
Experience in Reinforcement Learning (RLHF, RLAIF) and LLM post-training techniques (SFT, DPO, PPO).

Preferred qualifications:

Master’s degree or PhD in Engineering, Computer Science, or a related technical field.
8 years of experience with data structures and algorithms.
3 years of experience in a technical leadership role leading project teams and setting technical direction.
Experience in multimodal learning or embodied agents, integrating various signals (text, audio, vision) into unified reasoning models.
Experience building efficient evaluation harnesses, benchmarks, or simulation environments for measuring agent performance.
Proven track record (publications or production launches) in reward modeling, including dense/mixture of experts (MoE) architectures and hybrid reward systems.