Machine Learning Systems Engineer, Research Tools

<div class="content-intro"><h2><strong>About Anthropic</strong></h2> <p>Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.</p></div><div> <div> <h2>About the Role:</h2> </div> <div> <p>We are seeking an experienced Machine Learning Systems Engineer to join our Encodings and Tokenization team at Anthropic. This cross-functional role will be instrumental in developing and optimizing the encodings and tokenization systems used throughout our Finetuning workflows. As a bridge between our Pretraining and Finetuning teams, you'll build critical infrastructure that directly impacts how our models learn from and interpret data. Your work will be foundational to Anthropic's research progress, enabling more efficient and effective training of our AI systems while ensuring they remain reliable, interpretable, and steerable.</p> </div> <h2><strong>Responsibilities:</strong></h2> <ul> <li>Design, develop, and maintain tokenization systems used across Pretraining and Finetuning workflows</li> <li>Optimize encoding techniques to improve model training efficiency and performance</li> <li>Collaborate closely with research teams to understand their evolving needs around data representation</li> <li>Build infrastructure that enables researchers to experiment with novel tokenization approaches</li> <li>Implement systems for monitoring and debugging tokenization-related issues in the model training pipeline</li> <li>Create robust testing frameworks to validate tokenization systems across diverse languages and data types</li> <li>Identify and address bottlenecks in data processing pipelines related to tokenization</li> <li>Document systems thoroughly and communicate technical decisions clearly to stakeholders across teams