Anthropic Recruits Senior Technologists to Move into AI Safety: No ML Required, Starting at $350K a Year

Anthropic opened applications for the Anthropic Residency on November 2. Anthropic is the developer of Claude. This program is paid and fixed-term, lasting from 6 weeks to 6 months, and is aimed at professors, technical founders, and staff-level engineers. The official website is very direct: no machine learning background is required, no PhD is required, and no publications are required. Applications close on November 22. U.S. roles and U.K. roles are separate. ML means machine learning.

Compensation is $350,000 to $500,000 per year equivalent. Location preference is the San Francisco office; shorter durations can be remote. The program provides mentors, compute, and self-study time. The goal is for Residents to be able to do full-time AI research by the end of the program, whether they stay at Anthropic or go elsewhere. Applicants may also be considered for the Anthropic Fellows Program.

Anthropic explains why it is opening this round: ML is a young field, and many senior researchers today originally came from physics, mathematics, neuroscience, or software engineering. The stronger models become, the more important research skills are setting an agenda and judging what is worth doing. Senior technologists often already have this skill set, and the Residency helps them fill in their understanding of frontier AI and AI safety. The most counterintuitive thing about this recruitment is that it does not require an ML background, yet it requires you to already be a senior technologist. The bar has not been lowered; it has just shifted to a different dimension: from 'has done ML' to 'has led complex technical work.'

This round continues the Anthropic Fellows Program. Fellows have worked on subliminal learning, the Petri auditing tool, persona vectors, and other work. Residents will be matched with Anthropic researchers on topics, each will have a mentor, and there will be regular one-on-ones. Research directions include scalable oversight, adversarial robustness and AI control, model organisms, model internals and interpretability, safeguards, AI welfare, frontier red teaming, and ML systems and reinforcement learning. The mentor list includes Sara Price, Ethan Perez, Evan Hubinger, Samuel Marks, Elana Simon, Belinda Li, Jack Lindsey, Nicholas Carlini, and Jascha Sohl-Dickstein.

The reactions under the job post are more informative than the job description. Someone saw 'no ML background required' and said it was Anthropic's most targeted sentence; someone asked what people without an ML background can notice that researchers are too close to see; someone made a video staring at the salary; someone posted a cat staring at a MacBook, saying 'on it'; and someone else said they were using Claude to write a cover letter to protect Claude. Half the comments are self-deprecating; the other half are calculating the annual salary.

A user replied with this video of a cat sitting in front of a MacBook, captioned 'on it.' This probably best represents applicants' mood: first open the computer, then think about what ML is.

Seeing the salary starting at $350,000, a user posted a video of a TV show screenshot. A reply below directly says, '350-500k, not bad.'

Application entry is on Anthropic's official website: https://alignment.anthropic.com/2026/anthropic-residency/ . U.S. role: https://job-boards.greenhouse.io/anthropic/jobs/5438311008 . U.K. role: https://job-boards.greenhouse.io/anthropic/jobs/5439169008 . Deadline November 22. The official website also says applicants from different backgrounds are welcome, and it especially encourages applications from underrepresented groups in the tech industry.