Google has 584 active AI-related job listings. The majority of these roles are focused on agents, representing 40% of the total, and serving infrastructure, at 26%. The most frequent technical tags include model_serving, agent_orchestration, and evals. Over the last 30 days, Google has added 413 new AI roles, a 105% increase compared to the preceding 30-day period.
Currently tracking 498 active AI roles, down 12% versus the prior 4 weeks. Primary focus: Agent · Engineering. Salary range $98k–$1030k (avg $233k).
Google currently has 586 active AI-related roles in our index. The most common open titles are: Software Engineer (5), AI Adoption Customer Engineer, Google Cloud (3), Conversational AI Consultant (2), Engineering Manager, Egregious Abuse Protection (2), Forward Deployed Engineer III, Generative AI, Google Cloud (2). Most positions are in Engineering and Product.
Google's active AI hiring is concentrated in: agents (43%), serving infrastructure (25%), application (19%). These categories follow a seven-stage AI lifecycle: data, pre-training, post-training, serving infrastructure, agents, evaluation, and application.
Google is hiring AI talent in: United States (376 roles), India (53 roles), Singapore (40 roles), Switzerland (20 roles).
Job postings at Google most frequently mention: Software Engineering, Algorithms & Data Structures, System Design, Computer Architecture, Machine Learning.
In the past 30 days, Google has posted 571 new AI-related roles. That is a +22% change versus the prior 30 days (469 → 571).
| Title | Stage | AI score |
|---|---|---|
| Research Scientist, Generative AI, DeepMind Research Scientist at Google DeepMind focused on designing and developing novel generative methodologies, particularly diffusion models, for media synthesis and scientific discovery. The role involves collaborating with international teams, utilizing advanced deep learning techniques, and contributing to the advancement of AI for public benefit and product innovation. | Post-trainPretrain | 10 |
| Threat Modeler Lead, CBRNE, DeepMind Lead threat modeler for AI safety in CBRNE domains, focusing on evaluating and mitigating dual-use risks of advanced AI models. This role involves refining threat modeling frameworks, designing evaluations for AI risks, collaborating with mitigation teams, and engaging with external stakeholders. Requires a PhD and experience in national labs or defense organizations, with a preference for experience in red-teaming LLMs and understanding CBRNE risks. |
| Eval Gate |
| 9 |
| Research Scientist, Verified Code Generation, DeepMind Research Scientist role focused on developing AI systems for verified code generation using LLMs and formal methods, with a focus on programming language semantics and static analysis. | ShipData | 9 |
| Research Engineer, Frontier Safety Mitigations, DeepMind Research Engineer focused on frontier AI safety mitigations, defending against misuse domains like CBRNE and Harmful Manipulation. Responsibilities include building evaluations, red-teaming, deploying in-model and out-of-model mitigations, and monitoring risks for frontier models, particularly agentic AI systems. The role involves developing classifiers, monitoring systems, and advancing research in automated red-teaming and adversarial robustness. | AgentEval Gate | 9 |
| Research Engineer, Multi Agent Learning, DeepMind Research Engineer at Google DeepMind focused on developing novel multi-agent learning algorithms and frameworks. The role involves building and maintaining large-scale simulation platforms and research pipelines on cutting-edge infrastructure, partnering with Research Scientists to translate research into production-quality code, and optimizing the research workflow. Requires experience in deep learning frameworks, Python/C++, and distributed training on accelerators. A PhD in ML, RL, or Multi-Agent Systems and experience with language models are preferred. | AgentPost-train | 9 |
| Research Engineer, Responsibility Engineering, DeepMind Research Engineer at Google DeepMind focused on AI safety, developing post-training strategies to mitigate adversarial risks and building evaluation infrastructure for frontier language models. The role involves prototyping scalable engineering solutions, optimizing training and inference pipelines, and collaborating with research scientists to translate safety research into implementations. | Post-trainServe | 9 |
| Research Software Engineer, Generative AI Research Software Engineer focused on developing foundational models and core technologies for synthesizing reality, particularly human body, face, and related components, to power machine learning, build better products, and enable next-generation user experiences, with applications in AR and XR devices. The role involves developing algorithms for 3D body shape estimation, rigging, skinning, and physics-based generative animation conditioned on multimodal inputs, with a requirement for publication in AI conferences. | Post-trainServe | 9 |
| Research Engineer, Frontier Safety Mitigations, DeepMind Research Engineer focused on building safety mitigations for frontier AI models, defending against misuse in domains like CBRNE and Harmful Manipulation. Responsibilities include building classifiers, data pipelines, monitoring systems, and evaluating/securing agentic AI systems, with a strong emphasis on automated red-teaming and adversarial robustness research. | AgentEval Gate | 9 |
| Research Scientist, Language, DeepMind Research Scientist at Google DeepMind focusing on groundbreaking research in language technology, particularly multilingual and multicultural ability. The role involves solving new problems, improving existing models, developing technical solutions, and communicating research findings. Requires a PhD in NLP/ML or equivalent, Python, neural network training experience, and publication submissions. Preferred experience includes LLM pretraining, post-training, inference with multilingual data, and novel evaluations. | PretrainPost-train | 9 |
| Research Scientist, Protein Design, DeepMind Research Scientist at Google DeepMind focused on using generative machine learning models to design proteins with novel functions for wet-lab testing. The role involves developing models for in silico predictions of protein functions, troubleshooting design failures, and identifying research directions in protein design method development. Requires a PhD in a relevant field and experience with wet-lab experimental procedures and protein design tools. | Data | 8 |