IMPLICATOR
Job seekers who landed offers at Anthropic or OpenAI spent an average of $4,600 preparing, interviewing.io founder Aline Lerner told Bloomberg Businessweek in a feature published May 28, paying $170 to $550 an hour for mock interviews. The decisive round, by the coaches' own account, is the one that involves no coding at all. Anthropic calls it the culture interview, and the prep platforms that coach applicants say it rejects more strong candidates than any technical test.
That is the shape of getting hired at the world's most valuable AI startup. Anthropic puts candidates through as many as five rounds of interviews and skills assessments, and recruiters say the most candidates fail at the nontechnical one. Clearing every coding test buys a candidate the right to be judged on something $4,600 of coaching can rehearse but not guarantee.
Key Takeaways
- Candidates who land Anthropic or OpenAI offers spend an average of $4,600 preparing, much of it aimed at one round that involves no coding.
- The culture interview, a nontechnical hour, rejects more strong candidates than any technical round, recruiters say. Pre-packaged STAR-story scripts are the top failure mode.
- Anthropic bans AI tools in live interviews and sends candidates its safety writing to read before the onsite.
- Passing every round is not an offer. Candidates stall for weeks at team matching, then face a 12-month reapply wait with no feedback.
AI-generated summary, reviewed by an editor. More on our AI guidelines.
What the culture round asks
The values round is a standalone hour with non-technical interviewers, and it sits in the onsite for nearly every role, from research scientist to payroll specialist. Candidates describe it as closer to a therapy session than a job interview, deeply personal and emotionally probing. Kevin Landucci, a Bay Area coach who works for the prep service Exponent, told Bloomberg that applicants "liken it to an intrusive conversation that doesn't feel like it's within the bounds of a work conversation."
The questions are reflective and hypothetical rather than technical. Daniela Amodei, Anthropic's president and co-founder, described the round on a podcast last year as asking "what are some unusual beliefs that you hold and how have you defended those beliefs in kind of uncomfortable situations." Landucci coaches candidates to surface a past decision that "didn't sit well" with them, and says the company "actually want[s] you to be skeptical" of how it pursues its mission. Interviewing.io, citing Anthropic recruiters, tells users the round rewards pushback over rehearsed enthusiasm.
The round does not reward a binder of prepared stories, either. Ridhima Khurana, who wrote up her own culture interview, warned that walking in "with four pre-packaged STAR stories" and trying to "shoehorn them into whatever gets asked" is the fastest way to fail, a failure mode she says Anthropic flags directly. Khurana notes that Anthropic describes itself as high-trust and low-ego, and says the round is where the company pressure-tests whether a candidate fits that line.
| Stage | What it screens for | Where candidates stall | What the prep guides advise |
|---|---|---|---|
| Recruiter call 30 min | Mission alignment, why Anthropic specifically | Strong engineers who cannot say why Anthropic | Learn the safety mission and B Corp framing |
| Coding assessment CodeSignal, 60-90 min | First-principles coding and concurrency; skipped for some referrals | Over-rehearsed, pattern-matched answers | Talk through the code out loud; no AI tools allowed |
| Onsite loop 4-5 hours | Two coding rounds and system design tied to LLM serving | Memorized templates over visible reasoning | Read the engineering blog; expect a project deep-dive |
| Culture / values round 1 hour, non-technical | Whether you hold your values under pressure | The most rejections happen here | Do the safety reading; bring real ethical friction, not scripts |
| Team matching | Fit to a team with open headcount | Candidates who passed everything stall on seats | Rank team preferences; expect weeks of silence |
| Offer, or a 12-month wait | The committee's final call | Rejection comes with no feedback and a year before reapplying | Reapply earlier only if something material changed |
Where the $4,600 goes
The reading comes first. Recruiters send candidates Anthropic's safety writing before the onsite and expect formed opinions, and the prep guides point applicants to Dario Amodei's "Machines of Loving Grace" from 2024, his January 2026 essay "The Adolescence of Technology," and the 2023 "Core Views on AI Safety." Beyond the reading, the spend goes to rehearsal. Mock interviews on interviewing.io start around $179 a session and run past $300 for a company-specific interviewer, with three-session coaching packages near $2,000. At the $170-to-$550 hourly range Lerner cited, the $4,600 average covers roughly 8 to 27 hours of practice, or close to one percent of the $420,000-to-$746,000 median total compensation that Levels.fyi reports across Anthropic roles.
The AI business brief, every morning
Strategic AI news from San Francisco. No hype, no "AI will change everything" throat clearing. Just what moved, who won, and why it matters. Daily at 6am PST.
No spam. Unsubscribe anytime.
Anthropic's own candidate guidance tells applicants to "be yourself" and bans AI help in live interviews, where there is "no AI assistance unless we indicate otherwise" so the company can see how they think in real time. One prep writeup warned that vague enthusiasm reads as coached and specific stories read as real. The coaching sells a way to sound as though none of it was coached.
Team matching, and the weeks of silence
Passing every round does not produce an offer. Candidates who clear the onsite move to team matching, where Anthropic places a hire into a pool and circulates the profile to three to five team leads with open headcount. The stage can run two to four weeks, sometimes six to eight, with little communication. A former Anthropic recruiter told the careers site Leonstaff the silence is deliberate, "because we don't want to give false timelines that we can't meet."
People involved in Anthropic's hiring have said on forums that final rejections often turn on a shortage of seats or subjective tiebreaks rather than performance. One Blind poster, in an account prep researchers label unverified, described an interviewer fixating on a "midwest public university" background while noting Stanford and Harvard pedigrees. The hiring also runs ahead of the mission framing. In a late-May snapshot of the job board, sales was the largest category, with roughly 74 open roles against about 69 in research and engineering, and Leonstaff, citing Glassdoor data, puts the average decision near 20 days, from three for sales roles to 90 or more for research.
Anthropic has added about 1,000 people since November and now employs more than 3,000, a scale it reached while passing OpenAI at a $965 billion valuation and drawing hires like Andrej Karpathy. Candidates it turns away after the final round are told to wait 12 months before reapplying, longer than the six-month post-onsite cooldown one prep guide still cites, and the company gives no feedback on the decision. A second attempt means waiting that year and paying for prep again, with no guarantee the nontechnical round goes better. The $850,000 pay ceiling is why candidates keep trying.
Know someone who'd find this useful? ✉️ Email it to a friend in one click, or they can subscribe free here.
Frequently Asked Questions
What does Anthropic's culture interview ask?
Reflective and hypothetical questions about your values, ethical friction at work, and unusual beliefs you have defended. It tests whether you hold your values under pressure, not whether your opinions are correct. Walking in with pre-packaged STAR stories is the most common way to fail.
How many interview rounds does Anthropic have?
A recruiter call, a CodeSignal coding assessment, then a 4-5 hour onsite that includes a hiring manager call, two coding rounds, a system design round, and the values interview. Candidates can face as many as five rounds. AI tools are banned unless an interviewer permits them.
How much does it cost to prepare for an Anthropic interview?
Candidates who landed offers at Anthropic or OpenAI spent an average of $4,600, interviewing.io founder Aline Lerner told Bloomberg. Mock interviews run $170 to $550 an hour, starting near $179 a session, with three-session coaching packages around $2,000.
What is team matching at Anthropic?
After passing the onsite, you enter a pool and get matched to a team with open headcount rather than hired into a fixed role. It can take two to four weeks, sometimes six to eight, with little communication. Candidates can be turned away here for a shortage of seats, not performance.
Can you reapply to Anthropic after a rejection?
Yes. Anthropic asks rejected candidates to wait 12 months before reapplying, or sooner if something material has changed in their experience. The company does not give feedback on interview or team-matching rejections.
AI-generated summary, reviewed by an editor. More on our AI guidelines.
Marcus Schuler
Editor-in-Chief and founder of Implicator.ai. Former ARD correspondent and senior broadcast journalist with 10+ years covering tech. Writes daily briefings on policy and market developments. Based in San Francisco. E-mail: editor@implicator.ai
