Principal Applied Scientist, Unity

• This role is ideal for someone excited to explore new frontiers in multimodal modeling, grounded reasoning, and AI-assisted content creation
• Advanced degree (MS or Ph.D.) in Computer Science, Machine Learning, Robotics, or a related field - or equivalent practical experience
• Deep expertise in multimodal LLMs, spatial reasoning, world models, or closely related areas
• Proven track record of research impact, through publications or applied contributions to production systems
• Hands-on experience building and evaluating large models, particularly those grounded in perception or environment-based reasoning
• Proficiency in Python and experience with modern ML frameworks such as PyTorch or TensorFlow
• Strong understanding of environment-agent interaction, simulation frameworks, or embodied AI
• Ability to translate research ideas into high-quality, practical systems
• Excellent communication skills and experience collaborating with cross-functional teams including engineers, designers, and product managers
• Experience integrating AI research into game engines or real-time simulations
• Background in reinforcement learning, planning, or agentic systems

Desirable

• We’re looking for a Principal Applied Scientist to lead this work. In this role, you'll focus on advancing multimodal LLMs, strengthening spatial reasoning, and building world models that help creators author rich, interactive experiences
• Your research will directly influence how agentic systems are built inside the Unity engine and how developers bring their ideas to life through more intuitive AI-driven workflows
• This role is ideal for someone excited to explore new frontiers in multimodal modeling, grounded reasoning, and AI-assisted content creation
• Drive initiatives to improve spatial reasoning in large models, including understanding scene layouts, object relationships, navigation, and 3D dynamics
• Develop and refine world modeling approaches that allow AI systems to build, maintain, and reason over structured representations of complex game environments
• Advance multimodal LLM architectures that integrate text, vision, and spatial signals to support agentic game authoring and interactive creation tools
• Prototype new AI-assisted workflows, such as natural-language scene editing, spatial instruction following, or agent behavior specification
• Collaborate with engineering and product teams to translate research innovations into production-ready features and tools
• Define evaluation strategies and benchmarks grounded in real game-development scenarios to ensure models are robust, reliable, and performant
• Contribute to the long-term vision for Unity’s agentic AI ecosystem, balancing usability, scalability, and creator empowerment
• Mentor scientists and engineers while fostering a collaborative, innovative, and inclusive technical culture