Anthropic Careers: Teams, Locations and How Hiring Works
A machine learning engineer reads an Anthropic job ad for a role working on Claude's API. The title looks right, the stack matches, and the work sounds interesting. Then the location line says "Remote-Friendly (Travel-Required)" next to San Francisco, Seattle and New York, and the location policy further down expects staff in an office at least 25% of the time.
TL;DR
- Anthropic hires across research, engineering, product, policy, safety, operations, and go-to-market teams.
- Most roles are hybrid at an office hub. When we checked in October 2026, about 1 in 10 of Anthropic's 682 openings was marked remote-friendly, nearly all in the US.
- Competition is strong. Applications that show relevant AI, infrastructure, safety, or policy experience tend to do better.
- Pay and interview stages vary by seniority, location, and team. Check each listing before you assume global remote.
- JobsBoard tracks fresh listings for Anthropic and similar companies, so you can monitor new roles without checking multiple pages.
The Short Answer
Anthropic hires remotely for a small share of roles. Its postings say staff are expected in one of its offices at least 25% of the time, and some roles ask for more. The remote-friendly exceptions cluster in compute and data center work, security and public policy, and are mostly open to people in the US. Each posting states its own location, so treat every listing as its own rule.
| Team area | Typical roles | What they look for |
|---|---|---|
| Research and safety | Research scientist, alignment researcher, policy researcher | Strong machine learning or policy research background, clear writing, empirical rigor |
| Engineering | Software engineer, infrastructure engineer, security engineer, full stack engineer | Production systems experience, distributed systems skill, interest in AI systems and API reliability |
| Product and design | Product manager, product designer, developer experience | Technical product sense, experience with APIs or developer tools, user-centered thinking |
| Policy, safety, communications | Public policy, communications, trust and safety | Policy knowledge, safety judgment, ability to translate technical work for external audiences |
| Operations and go-to-market | Legal, finance, people, recruiting, sales, partnerships | Scaling company operations, compliance exposure, enterprise or developer ecosystem experience |
What the Company Does and Which Teams Hire
Anthropic describes itself as an AI safety and research company. It builds large language models and AI systems, with Claude as its main product. The company also publishes research on model behavior, alignment, and safety, so its open roles go beyond standard software engineering.
Anthropic hires researchers, engineers, product people, policy staff, and the operational teams that support them. The work spans model development, API products, safety research, security, developer experience, and enterprise adoption. This creates a broader set of open roles than some people expect. A backend engineer, technical writer, policy researcher, or sales leader may find a relevant opening. If you're comparing stacks, roles for full stack work are covered in Remote Full Stack Developer Jobs: Which Stacks Get Hired and Where. Developer ecosystem and developer relations adjacent roles also exist across companies like Anthropic, and the expectations can vary. For a broader look at how those roles are defined, see Developer Relations Jobs: What DevRel Teams Hire For.
Remote Setup and Where It Hires From
Anthropic's default is hybrid. Postings carry a location-based hybrid policy of at least 25% office time, a few roles say 75%, and many remote-friendly roles add that travel is required. Even postings marked remote-friendly include that policy line, so ask the recruiter early how much office time the role really needs.
Candidates outside the US should not assume every listing is open to them. Outside the US, roles are mostly tied to offices such as London, Dublin, Munich, Tokyo, Seoul and Singapore; when we checked, the only remote-friendly roles outside the US were a handful in Canada and one in Australia. Anthropic's postings say it does sponsor visas, but not for every role or every candidate. Remote work and employment status also vary by country, and tax or immigration rules are not the same across borders. Before acting on a specific role, confirm eligibility with the company and, where needed, an immigration or tax professional. If you're in the UK, remote AI work is not limited to one employer. You can also compare Remote Jobs in the UK: Tech Roles You Can Do From Anywhere in Britain for a broader look at UK-eligible roles.
JobsBoard groups remote positions by industry and region, which helps you avoid scrolling through roles that are not open where you live. You can sweep through remote AI jobs to see which teams are hiring in your region right now. Filter by department and role once you have a clearer target.
How Hiring Generally Works There
For most posted roles, expect a process that starts with an application review and recruiter conversation, then a technical or work-sample stage, followed by conversations with the hiring team. The exact loop varies by team and seniority. Public postings often emphasize careful thinking about safety, product tradeoffs, and research judgment.
You likely won't get one standard interview loop for every job. A research scientist may do research talks and technical discussions. A product manager may do case-based interviews and product thinking exercises. A software engineer may do coding, system design, and infrastructure discussions. Some roles ask for work samples or written exercises. Because the process can be slow, especially for research and senior positions, apply early and keep a focused pipeline rather than waiting on one posting.
For hiring managers creating similar roles, making the scope, location, and decision authority clear early helps candidates self-select. The same principles in How to Write a Job Description for AI, Web3 and Software Roles apply to postings in this space. If you're hiring on the other side, you can post a job on JobsBoard to reach candidates who are already looking across AI, Web3, and software tools.
How to Stand Out
The strongest candidates show specific evidence of building, researching, or operating in adjacent technical spaces. For engineering, that means production systems, API design, infrastructure, or security work. For policy and safety roles, that means writing clearly about tradeoffs and understanding how AI systems behave outside a demo.
Anthropic's postings ask people to apply even if they don't meet every listed qualification, so a near miss is worth a try. Generic applications still tend to do poorly. Tailor your resume and cover note to the team and problem area. If a role mentions API reliability, highlight incidents you diagnosed or systems you hardened. If a role mentions safety evaluation, describe how you measured model behavior or built evaluation tooling. Every posting also links Anthropic's guidance on candidates' AI use, so read it before you lean on AI tools for the application. Use the free resume match to compare your resume against live listings before you spend time on a specific application. Then set free weekly job alerts for Anthropic and adjacent companies so new roles land in your inbox instead of requiring daily checks.
If you're weighing AI safety careers against other technical roles, AI-Proof Jobs: Which Roles AI Will Replace and Which It Won't is a useful companion read. It covers how different job categories are changing as AI tools spread. No credential or single project will guarantee an interview, but a crisp signal about what you can do in the role usually outperforms a broad list of skills.
Frequently Asked Questions
Is Anthropic fully remote?
No. Postings expect staff in an office at least 25% of the time, and only about 1 in 10 roles was marked remote-friendly when we checked, mostly in the US. Research roles are the most office-bound: 8 of 67 AI research and engineering openings were remote-friendly.
What jobs can I get at Anthropic without a PhD?
Software engineering, product management, design, policy, operations, legal, and go-to-market roles often do not require a PhD. Research and alignment scientist roles frequently expect a PhD or equivalent research experience. Where postings mention a degree, they usually accept an equivalent mix of education, training and experience.
Does Anthropic hire outside the US?
Yes, at offices including London, Dublin, Munich, Tokyo, Seoul and Singapore, but remote roles outside the US are rare (a few in Canada and Australia when we checked). Anthropic says it sponsors visas, though not for every role. Verify eligibility and any visa or tax implications before applying.
How hard is it to get a job at Anthropic?
Competition is strong for research and engineering roles, but there is no fixed acceptance rate that tells you your odds. A focused application with relevant evidence of production, research, policy, or operational work is more useful than applying to many roles with the same generic resume.
Does Anthropic pay remote workers differently?
Anthropic's postings list an annual pay range for the role, and for sales roles that range is on-target earnings including commission. Pay at AI labs varies by location, seniority and team, so compare the posted range with your location and ask the recruiter once you're in process.
Where to Look Next
Start with Anthropic's open roles on JobsBoard and filter by your region and department. If nothing fits today, check companies hiring now to see other AI, Web3, and software tools companies with fresh listings. JobsBoard keeps thousands of remote and global job listings across those three industries, and closed roles are taken out of search rather than left up to collect applications.
