New: upload your resume and get matched to AI jobs. Free to start →

Senior Research Engineer, Embodied Generalist Agent, Gaming

Tokyo, Japan • Full-time

Posted 1mo ago

Remote Work Policy

On-site

Employment Type

Full-time

Categories

AI Agent Engineer

About the job

DeepMind is seeking a highly motivated and innovative Research Engineer in Tokyo to build multimodal embodied gaming agents. You will collaborate with researchers and engineers to develop agents capable of perceiving, reasoning, planning, and executing real-time actions in complex environments. This role involves utilizing the latest advancements in multimodal LLMs, VLA models, ICL, SFT, and RL to solve fundamental challenges in embodied intelligence, bridging high-level planning with low-level motor control. You will contribute to creating adaptive agents that master tasks in rich virtual testbeds, pushing the frontiers of AI agents and defining the next era of adaptive intelligence.

Responsibilities

  • Develop and optimize agent architectures for multimodal perception, reasoning, and real-time execution.
  • Build and scale training recipes using supervised fine-tuning, reinforcement learning, imitation learning, or in-context learning.
  • Design systems for agents to reason over long horizons and utilize memory for complex tasks.
  • Research and implement capabilities for agents to adapt to new environments and learn at test time.
  • Establish benchmarks in virtual environments to measure progress in general agent capabilities and embodied intelligence.

Requirements

  • Bachelor’s degree or equivalent practical experience.
  • 5 years of experience with software development in one or more programming languages.
  • 3 years of experience testing, maintaining, or launching software products.
  • 1 year of experience with software design and architecture.
  • Experience with ML frameworks such as JAX, TensorFlow, or PyTorch.
  • Experience building agents for 3D virtual environments, simulators, or video games (preferred).
  • Solid understanding of LLM internals, Deep Reinforcement Learning (RL), LLM Reasoning, imitation learning, memory-based architectures, VLM, or VLA models (preferred).
  • Proven track record of designing, implementing, and maintaining robust technical assets for technical stakeholders (preferred).

About Google DeepMind