How to Get Referred to Anthropic or OpenAI Without Connections

Landing a role at a frontier AI lab like Anthropic or OpenAI feels like trying to get into an exclusive club where you don't know the bouncer. These companies receive tens of thousands of applications for every open role, and the competition is fierce. But here's the thing most candidates miss: the single most effective way to stand out isn't a fancier resume or a shinier portfolio. It's a referral from someone already inside.

Referred candidates are significantly more likely to get interviews and receive offers compared to cold applicants. That's not a secret. What feels like a secret is how to actually get one of those referrals when you don't know a single person at Anthropic, OpenAI, or any other AI lab. The good news? You don't need a pre-existing network. You need a strategy.

Whether you're a software engineer looking to break into AI safety research, an ML engineer eyeing a position on a foundation model team, or a backend developer who wants to build infrastructure that powers the next generation of AI products, this guide will walk you through exactly how to earn a referral at the companies shaping the future of artificial intelligence. And if you want to skip the cold outreach grind entirely, you can browse Anthropic referral opportunities on ReferMe and connect directly with employees who are ready to refer qualified candidates.

Let's break down the playbook.

Why Referrals Are the Golden Ticket at AI Labs

Before diving into tactics, it helps to understand why referrals carry so much weight at companies like Anthropic and OpenAI specifically.

AI labs operate differently from most tech companies. Their teams are small relative to their impact, which means every hire matters enormously. A bad hire on a six-person research team is far more disruptive than a bad hire on a two-hundred-person engineering org at a Fortune 500 company. Because the stakes are high, hiring managers lean heavily on internal signals. When an existing employee vouches for someone, it tells the hiring team: "I've evaluated this person, and I'm putting my own reputation on the line to say they'd be a great fit."

The numbers back this up across the tech industry. Referred candidates typically account for a disproportionate share of hires relative to their share of total applications. At smaller, high-bar organizations, this effect is even more pronounced. According to the Bureau of Labor Statistics, computer and mathematical occupations, which include most AI lab roles, are among the fastest-growing job categories. That growth means more applicants flooding the pipeline, which in turn makes referrals even more valuable as a filtering mechanism.

At Anthropic and OpenAI, the hiring process is rigorous. Both companies evaluate not just technical skill, but alignment with their mission, ability to collaborate across disciplines, and capacity for independent problem-solving. A referral doesn't guarantee you'll pass all of those bars, but it almost always guarantees that a real human being will look at your application instead of an automated system filtering you out.

Here's what a referral actually does behind the scenes at most AI labs:

• Bypasses initial resume screening. Your application gets flagged and routed to a recruiter or hiring manager directly.

• Adds social proof. The referrer's internal reputation transfers partially to you.

• Provides context. A good referral includes a note explaining why you're a fit, something a resume alone can't convey.

• Accelerates the timeline. Referred candidates often move through interview stages faster because there's already a baseline of trust.

So the question isn't whether you should pursue a referral. It's how to get one when your LinkedIn connections page doesn't include anyone with "@anthropic.com" or "@openai.com" in their profile.

The answer lies in a combination of strategic relationship-building, visible contribution to the AI community, and using platforms designed to connect job seekers with referrers. You can also explore OpenAI referral opportunities on ReferMe to see which roles have employees actively offering referrals right now.

Building Visibility Before You Ask for Anything

The biggest mistake engineers make when trying to get referred to a top AI lab is leading with the ask. Imagine receiving a message from a stranger that says, "Hi, I see you work at Anthropic. Can you refer me?" You'd ignore it. Everyone does.

Instead, you need to become someone worth referring before you ever send a single outreach message. This doesn't mean you need a PhD from Stanford or three ICML papers (though those help). It means you need to be visible, credible, and easy to vouch for.

Contribute to Open Source AI Projects

Both Anthropic and OpenAI have a strong culture of open research and open source tooling. Contributing to projects in their ecosystem, or adjacent to it, is one of the most direct ways to get noticed by the people who work there.

Some concrete starting points:

• Anthropic's ecosystem: Look at projects related to constitutional AI, interpretability tooling, or the Claude API developer tools. Even filing well-documented issues or improving documentation counts as meaningful contribution.

• OpenAI's ecosystem: Contribute to libraries like Tiktoken, Evals, or Whisper. If you've built something interesting using the OpenAI API, write about it publicly.

• Adjacent tools: Projects like LangChain, LlamaIndex, vLLM, and Hugging Face Transformers are deeply connected to the work happening at these labs. Active contributors to these repos are on the radar of AI lab engineers whether they realize it or not.

When you contribute publicly, you create artifacts that a potential referrer can point to. "I've seen this person's PRs on [X project], and their code quality is strong" is infinitely more compelling than "someone I've never heard of asked me for a referral."

Write About What You're Learning

You don't need to publish groundbreaking research. You need to demonstrate that you think clearly about hard problems. Blog posts, technical write-ups, or even detailed Twitter/X threads about topics relevant to AI lab work can put you on the map.

Topics that resonate with people at Anthropic and OpenAI include:

• Scaling challenges in large language model infrastructure

• Evaluation methodology for AI systems

• Safety and alignment considerations in deployment

• Performance optimization for inference workloads

• Developer experience improvements for AI APIs

One well-written blog post about how you optimized a transformer inference pipeline, complete with benchmarks and code, can do more for your career than hundreds of cold DMs.

Engage Authentically in AI Communities

Slack communities, Discord servers, local meetups, and online forums are where AI lab employees hang out when they're not at work. The Alignment Forum, EleutherAI's Discord, and ML-focused subreddits are all places where you can build genuine relationships over time.

The key word is genuine. Don't join a community with the sole purpose of extracting a referral. Join because you're genuinely interested in the problems these companies are solving. Ask thoughtful questions. Share useful resources. Help others. Over weeks and months, you'll naturally build connections with people who happen to work at the companies you're targeting.

The Outreach Playbook That Actually Works

Once you've built some visibility, it's time to reach out to specific people. This is where most candidates fumble, so let's be precise about what works.

Step 1: Identify the Right People

Not everyone at Anthropic or OpenAI is equally likely to refer you. You want to target people who:

• Work on the team or in the function you're applying to

• Have been at the company long enough to have credibility (six months or more)

• Are active on social media or in public communities (this signals they're open to connecting)

• Have something in common with you (same university, same previous employer, shared interests)

LinkedIn is the obvious starting point, but don't overlook GitHub (check who's committing to the company's public repos), Twitter/X (search for employees discussing their work), and conference speaker lists.

Step 2: Craft a Message Worth Responding To

Your outreach message should be short, specific, and demonstrate that you've done your homework. Here's a template that works:

Hi [Name], I've been following your work on [specific project or post they shared]. I'm a software engineer focused on [your specialty], and I recently [specific accomplishment relevant to their work, like a blog post, open source contribution, or project]. I'm really interested in the [specific team] at [Anthropic/OpenAI] and would love to learn more about what it's like working there. Would you be open to a 15-minute chat?

Notice what this message does: it shows you know who they are, demonstrates your own credibility, and makes a small ask (a conversation, not a referral). The referral comes later, after they've had a chance to evaluate you.

What this message does NOT do: ask for a referral upfront, send a resume as an attachment, or use generic flattery like "I admire your company's mission."

Step 3: Use a Referral Marketplace

Cold outreach works, but it's slow and unpredictable. You might send 20 messages and get two replies. That's normal, but it's also frustrating when you're motivated and ready to move.

This is exactly the problem that referral marketplaces solve. On platforms like ReferMe, employees at companies like Anthropic and OpenAI actively opt in to provide referrals. They're not strangers you're cold-messaging. They're people who have said, "Yes, I'm willing to refer qualified candidates." That changes the dynamic entirely.

When you sign up as a job seeker on ReferMe, you can browse open roles at AI labs, request referrals from verified employees, and even use AI-powered tools to tailor your resume for specific positions. It's the difference between knocking on doors in a neighborhood and walking into an open house.

If you're serious about getting into Anthropic or OpenAI, combining organic outreach with a structured referral platform gives you the best odds.

Step 4: Follow Up Without Being Annoying

If someone doesn't respond to your first message, send one follow-up after five to seven days. Keep it brief:

Hi [Name], just wanted to bump this in case it got buried. Totally understand if the timing doesn't work. If you're open to chatting at some point, I'd really appreciate it.

After that, let it go. Persistence is good. Pestering is not. If someone doesn't respond after two messages, move on to the next person on your list.

Preparing Your Application to Be Referral-Ready

Getting a referral is only half the equation. The other half is making sure your application is strong enough that the person referring you feels confident doing so. Nobody wants to put their name behind a candidate who submits a generic resume.

Tailor Your Resume for the Specific Role

AI labs care about different things than most tech companies. Your resume should highlight:

• Technical depth over breadth. Instead of listing 15 programming languages, go deep on the systems you've built and the problems you've solved.

• Scale and impact. Did you optimize a system that processes millions of requests? Did you reduce latency by 40%? Quantify everything.

• Relevant domain knowledge. If you're applying to an infrastructure role at Anthropic, emphasize distributed systems, GPU computing, or ML pipeline work. If you're targeting a research engineering role at OpenAI, highlight your experience with training runs, evaluation frameworks, or model deployment.

• Mission alignment. Both Anthropic and OpenAI care deeply about AI safety and responsible development. If you've done any work related to these themes, make it prominent.

ReferMe's AI resume builder can help you create an ATS-optimized resume tailored specifically for AI lab applications. It analyzes job descriptions and suggests how to frame your experience for maximum impact.

Prepare a "Referral Brief"

When someone agrees to refer you, make their job easy. Send them a short document (a few paragraphs) that includes:

• The specific role you're applying to (with a link to the job posting)

• A two-sentence summary of your background

• Two to three reasons why you're a strong fit for this particular role

• Any specific projects or accomplishments you'd like them to mention

This isn't pushy. It's respectful of their time. Most referrers appreciate having talking points because it means they can write a stronger internal recommendation.

Build a Portfolio That Speaks for Itself

Beyond your resume, AI labs want to see evidence of your thinking and your craft. Consider building:

• A GitHub profile with well-documented projects (clean READMEs, tests, clear commit histories)

• A personal site or blog with technical write-ups

• A portfolio of side projects that demonstrate relevant skills (fine-tuning models, building evaluation harnesses, creating developer tools)

The goal is to give your referrer ammunition. When they submit your name internally, they should be able to say, "Check out this project they built" or "Read this post they wrote about distributed training."

Getting referred to Anthropic or OpenAI without existing connections is absolutely possible. It requires effort, patience, and a willingness to put yourself out there before you ask for anything in return. Build visibility through open source contributions and public writing. Reach out to the right people with thoughtful, specific messages. Use referral platforms to connect with employees who are actively willing to help. And make sure your application materials are polished enough that anyone would feel proud to refer you.

The engineers working at these companies were once in your position. Most of them remember what it was like to be on the outside looking in, and many are genuinely happy to help strong candidates get their foot in the door.

Ready to take the first step? Browse Anthropic referral opportunities on ReferMe or explore open roles at OpenAI and connect with employees who can put your name forward. For more strategies on earning referrals from people outside your immediate network, check out our guide on how to get referred to a job without knowing anyone. Your next career move might be one conversation away.