Editor's Note: Anthropic is often imagined by the public as an AI lab composed of PhDs, researchers, and cutting-edge model experts. However, this breakdown of 1,680 engineer resumes provides a more realistic answer: Anthropic's core is not just about "research" but about "building."
This article analyzed 5,306 profiles on LinkedIn that listed Anthropic as their current workplace and further narrowed it down to 1,680 engineer resumes. The counterintuitive conclusion reached was that the most central figure at Anthropic is not the envisioned "researcher" but a group of experienced "builders" (individuals who can truly construct, run, and scale large-scale systems).
The data shows that Anthropic's engineering team has rapidly taken shape in the past 18 months: over half of the engineers have joined in less than a year, yet the new hires are generally very seasoned. The median work experience before joining is 12.2 years, with many from companies known for their engineering capabilities and infrastructure such as Google, Meta, Amazon, Microsoft, Stripe, Databricks, Snowflake, Palantir, and others.
This also explains the true focus of Anthropic's engineering organization: rather than focusing on model research as the public perceives, it is more akin to a highly engineered infrastructure company. The engineering backgrounds are mainly concentrated in infrastructure, backend, distributed systems, databases, and security. Only 13.7% hold a PhD, with the majority being senior engineers with undergraduate or master's degrees.
Early-career professionals are not entirely without opportunities, but the bar is equally high: internships at top tech companies, competition achievements, paper publications, or experience in AI safety/alignment projects often serve as signals in place of years of work.
The author's final advice is straightforward: if you want to join Anthropic, don't tailor your resume to look like you're applying to a research lab; instead, highlight the large-scale systems you have genuinely built, extended, and maintained. At the core of cutting-edge AI competition lies an increasingly intense competition of engineering and infrastructure capabilities.
Below is the original text:
Builders, Not Researchers
I scraped all LinkedIn profiles that list Anthropic as their current employer, totaling 5,306 individuals. I then filtered out the 1,680 individuals who truly hold engineering positions and further examined the 7,986 job records in their past job descriptions to analyze what they were doing before joining Anthropic.
Here are the results.
Organizational Expansion Almost Overnight
Only 15 engineers who joined Anthropic before 2021 and are still with the company. By 2025, the organization's engineering team had nearly tripled, with 686 engineers added that year; the hiring pace for 2026 is expected to be similar, with 455 new hires as of June.
Half of the current engineering team have been at Anthropic for less than a year. 53% of individuals joined in the past 12 months. Median tenure: 10 months.
This is a large-scale organization that was built in just about 18 months.
Almost Exclusive Hiring of Senior Engineers
The median work experience before joining Anthropic is 12.2 years. The middle 50% have between 8.8 and 16.5 years of experience. Out of these 1,680 individuals, only 50 have less than 3 years of work experience. 44% have 13 years or more of experience. Hiring of fresh graduates is virtually nonexistent.
In other words, a typical new Anthropic employee is an engineer with 12 years of experience but has only been at Anthropic for 10 months.
Clearly More Skewed Toward Infrastructure Than Traditional Research
An infrastructure background is present in 40% of engineers' resumes. Backend, distributed systems, databases, and security each account for about 20%. Reinforcement learning, the "RL" in RLHF, only appears in 3.3% of individuals' resumes.
A typical Anthropic engineer has typically spent the past decade at a hyperscale cloud vendor or a foundational infrastructure startup, building large-scale production systems.
Their self-reported skills also indicate the same: Python 585 people, Java 566 people, C++ 443 people, JavaScript 376 people, SQL 302 people, Linux 230 people, Distributed Systems 189 people, AWS 154 people. While there are certainly job roles related to more "sexy" topics like model training, they make up a small percentage.
The Biggest Talent Source is Not a Lab, But Google
Many people assume Anthropic primarily hires from OpenAI and DeepMind. However, its largest talent pipeline, by far, is Google. The labs of those competitors are merely two small bars in the middle of the chart.
Anthropic clearly prefers engineers known for their engineering rigor from companies like Stripe, Databricks, Snowflake, Palantir, and Airbnb.
Looking at the historical workplaces of these engineers, the ranking is: Google 405 people, Meta 273 people, Amazon 197 people, Microsoft 171 people, Stripe 124 people, Apple 87 people, Stanford 68 people, DeepMind 62 people, Airbnb 51 people, OpenAI 48 people. Currently, half of the engineering team, 50%, has at least one FAANG on their resumes.
Of course, they are also poaching talent from other AI labs. OpenAI is one of the top five direct sources, and DeepMind is one of the top six direct sources. Approximately 94 engineers have directly moved from other cutting-edge AI labs to Anthropic.
The Myth of the Ph.D.
Only 13.7% of individuals hold a Ph.D. That's roughly one in seven people.
Anthropic's typical recruit is not a research scientist but a senior engineer with a bachelor's or master's degree. The notion of "everyone in the lab has a Ph.D." is largely inaccurate at the engineering team level.
The distribution of professional backgrounds also aligns with the profile of a "building" organization: Computer Science 819 people, followed by Mathematics 78 people, Physics 70 people, Computer Engineering 69 people. Philosophy also makes it into the top 20, with a total of 13 people, likely related to the security domain.
Stanford Leads Significantly in Recruiting Sources
Looking at the school perspective, the historical cumulative ranking is as follows: Stanford 144 people, Berkeley 118 people, MIT 80 people, CMU 73 people, Harvard 42 people, Cambridge 39 people, UW 36 people, Waterloo and Cornell each 35 people, Oxford 33 people, Princeton 32 people. The top four schools together account for a quarter of the entire engineering team.
80% of the people hold the same job title.
They are all titled as "Member of Technical Staff."
A former Instagram CTO, several former Adept founders, and Stanford faculty all carry the title "MoTS" at Anthropic. This intentional flattening of job titles is evident. Qualifications and specific roles are deliberately obscured in the design.
For those early in their careers, what is the only pathway into Anthropic?
There are 172 engineers with less than 6 years of work experience, among them 50 have less than 3 years. However, they are not your typical entry-level candidates. They are roughly divided into two categories, with few ordinary mid-level engineers in between.
Compared to the entire engineering team, they exhibit markedly different characteristics: a higher percentage hold Ph.D. degrees, at 19%, compared to 13.7% for the overall team; the proportion of Product / SWE titles is three times that of the overall team, at 15%, compared to just 5% overall; their likelihood of having FAANG experience is much lower, at only 32%, compared to 50% for the overall team.
What they substitute for work tenure is another form of prestige capital:
Internship pipelines. Among them, 50% list internship experiences at the following companies: Meta 16 people, Google 10 people, DeepMind 6 people, Microsoft 5 people, Amazon 5 people, as well as Jane Street, Two Sigma, HRT, Optiver, Nvidia.
From quant trading to AI labs. 9% have passed through top trading firms, including Jane Street, Two Sigma, Five Rings, HRT, Optiver, Citadel. They are a group of young math / competitive programming talents who transitioned from high-frequency trading to AI labs.
Alignment Fellowship. 6% of people have had contact with MATS, SERI, Redwood, or ARC. This is an entrance that is almost exclusively open to early-career talent and virtually absent in the senior cohort.
A very clear picture is: MIT, IOI Silver, Codeforces 2900+ rating, directly entering reinforcement learning and security after four years of work. Their selection criteria are not years of work but rather competition rankings and paper publications.
These young engineers are also more international than senior engineers. The school backgrounds of junior engineers include: 15 from Berkeley, 14 from Stanford, 10 from Cambridge, 7 from MIT, 7 from Tsinghua, 6 from Oxford, in addition to Imperial, NUS, Shanghai Jiao Tong University, ETH Zürich.
So, how should you interpret this information?
If you want to join Anthropic as an engineer, do not make your resume look like it's for a research lab, but rather make it look like it's for an infrastructure company. Showcase systems you have actually built and scaled. That is the resume that is currently being hired.
The only exception is the early career stage. At this stage, the threshold is not regular work experience but top internships, competition rankings, or papers.
If you are competing with Anthropic for talent, your target is also not a "Ph.D." or "lab background" per se, but those senior Builders from hyperscale cloud vendors or highly reputable engineering companies: they have approximately 12 years of experience and may come from Stripe, Databricks, Snowflake, Palantir. Anthropic has been actively fishing in this talent pool.
Welcome to join the official BlockBeats community:
Telegram Subscription Group: https://t.me/theblockbeats
Telegram Discussion Group: https://t.me/BlockBeats_App
Official Twitter Account: https://twitter.com/BlockBeatsAsia
