Research Scientist/Engineer - Alignment Finetuning

Anthropic is seeking a Research Scientist/Engineer to join their Alignment Finetuning team, focusing on developing AI systems that are safe and beneficial for society. This role combines cutting-edge ML research with practical implementation, working on training language models that demonstrate better moral reasoning, improved honesty, and good character.

The position offers an opportunity to work with a collaborative team of researchers, engineers, and policy experts in a company that approaches AI research as an empirical science. You'll be responsible for developing novel finetuning techniques, implementing advanced training pipelines, and creating evaluation frameworks to measure and improve model alignment properties.

The ideal candidate will have advanced education in Computer Science or ML, strong programming skills (particularly in Python), and experience with ML model training and experimentation. The role requires both technical expertise and the ability to collaborate effectively across teams.

Anthropic offers competitive compensation ($280,000 - $425,000), flexible working arrangements, and a strong benefits package. The company values diversity and encourages applications from candidates with varied backgrounds and perspectives. As a public benefit corporation, Anthropic is committed to developing AI systems that are reliable, interpretable, and beneficial for society.

Working at Anthropic means joining a cohesive team focused on large-scale research efforts rather than smaller, specific puzzles. The company's research builds on significant work in areas like GPT-3, Circuit-Based Interpretability, and AI Safety, making this an excellent opportunity for those interested in advancing the field of beneficial AI.