Multimodal AI Researcher/Engineer

Anthropic is seeking a Multimodal AI Researcher/Engineer to join their team focused on creating safe and beneficial AI systems. The role combines research and engineering, working on frontier AI systems that operate across multiple modalities including images, video, and audio. The position involves developing new architectures, building infrastructure, and working with large-scale data processing.

The team operates across a comprehensive stack including training, inference, system design, and data collection. Key areas include foundational research in multimodal architectures, building complex reinforcement learning environments, developing high-performance servers, and creating tools for large-scale data processing.

Anthropic operates as a cohesive team focused on high-impact AI research, viewing it as an empirical science similar to physics and biology. The company values collaboration, communication, and a focus on advancing long-term goals of creating steerable, trustworthy AI. They have a strong track record in significant AI research including GPT-3, Circuit-Based Interpretability, and Multimodal Neurons.

The position offers competitive compensation (£250,000 - £270,000 GBP), comprehensive benefits, and a collaborative work environment in San Francisco. The role requires at least 25% office presence, with visa sponsorship available for qualified candidates. Anthropic particularly encourages applications from underrepresented groups and values diverse perspectives in addressing the social and ethical implications of AI systems.