At the last EAG Bay Area, I gave a workshop on navigating a difficult job market, which I repeated days ago at EAG London. A few people have asked for my notes and slides, so I’ve decided to share them here. 

This is the slide deck I used.  

Below is a low-effort loose transcript, minus the interactive bits (you can see these on the slides in the form of reflection and discussion prompts with a timer). In my opinion, some interactive elements were rushed because I stubbornly wanted to pack too much into the session. If you’re going to re-use them, I recommend you allow for more time than I did if you can (and if you can’t, I empathise with the struggle of making difficult trade-offs due to time constraints). 

One of the benefits of written communication over spoken communication is that you can be very precise and comprehensive. I’m sorry that those benefits are wasted on this post. Ideally, I’d have turned my speaker notes from the session into a more nuanced written post that would include a hundred extra points that I wanted to make and caveats that I wanted to add. Unfortunately, I’m a busy person, and I’ve come to accept that such a post will never exist. So I’m sharing this instead as a MVP that I believe can still be valuable –certainly more valuable than nothing! 

Introduction

80,000 Hours’ whole thing is asking: Have you considered using your career to have an impact?

As an advisor, I now speak with lots of people who have indeed considered it and very much want it – they don't need persuading. What they need is help navigating a tough job market.

I want to use this session to spread some messages I keep repeating in these calls and create common knowledge about the job landscape. 

But first, a couple of caveats:

1. Oh my, I wonder if volunteering to run this session was a terrible idea. Giving advice to one person is difficult; giving advice to many people simultaneously is impossible. You all have different skill sets, are at different points in your careers, and care about different problems. I'm going to throw ideas around, and it's up to you to decide which ones are most relevant to you. If you want tailored advice, you can apply to speak with me or one of my colleagues at 80k.
2. I'm giving my opinions as an 80k advisor. This is different from representing the opinions of 80k as an organisation. There are 30+ employees at 80k; we agree on many things, but we also have disagreements, and I'm only speaking for my opinionated self.

General Observations

Market inefficiencies

While job seekers often struggle to land jobs, organisations (including well-established ones) also sometimes struggle to hire. This can be confusing and frustrating for both sides.

Job seekers tell me: "What do you mean that  orgs are talent-constrained? I keep getting all of these rejection emails saying ‘sorry, we got hundreds of applications, it’s very competitive, don’t feel bad, bye’.” 

Meanwhile, some organisations ask me: "We hear there are many people looking for jobs... Hm, you work at 80k —do you know where these people are? They're certainly not applying for our jobs. Is something wrong with our job ad? Are we framing requirements incorrectly?"  [spoiler alert: sometimes there’s room for improvement in job posts and how they frame requirements, yes].

What's going on here? I'm not sure I have a great answer, but I have some hypotheses!

1. Some skill sets are genuinely in low supply. I'll say more about this shortly.
2. Many people aren't applying for jobs they should, because:
   • They don't know about them
   • They know about them but don't apply because it's a big time investment and emotionally taxing, and it doesn’t feel worth it.
   • They (wrongly) assume they aren't a good fit
   • They (wrongly) assume their comparative advantage is something else and focus on applying for other types of roles. 

On that last point: people don't always realise that your comparative advantage might be different from your absolute advantage. Sometimes it's better to work on something you're good at and everyone else is terrible at, rather than something you're excellent at, if everyone else is also excellent at it (and there aren't enough jobs for all the excellent talent).

For example, I've found that some people with STEM backgrounds may overfocus on technical research (like AI safety research) and don't consider other paths like policy, which is tremendously important at the moment and very much in need of technical people. I don't want to say that you specifically should be working on policy instead of technical research —I don't know you personally, and research is also very important. Use your own judgment! But as a general rule, I’ve found that technical people don't consider policy work (or comms work, or ops work) as often as they should.

I've also been guilty of not applying for jobs I should have. The majority of the job offers I've received over the last couple of years —THE MAJORITY— were for jobs I was aware of but not planning to apply for until the hiring manager reached out saying, "Hey, I don't know you, but someone suggested you might be a good fit. Have you considered applying?" Then I'd think, "OK, fine, if you say so, I'll apply."

This is not a robust system. It's pure luck that an organisation asked someone who happens to know and remember me. Don't be like me: don’t wait for people to reach out —you should apply proactively. But also, be like me: try to put yourself in a position where people have you on their radar and will nudge you to apply for things that you might disregard otherwise. 

Some Skills in High Demand and Low Supply

Some skills are particularly valued and in low supply. 

The main trend is that it's difficult to find candidates who have both relevant experience in these AND context on the cause area the organisation operates in (AI, biosecurity, meta-EA, etc.). I recommend this post to understand what "context" means, why it's important in hiring processes, and how you can level up. 

Specific skill sets that organisations find particularly difficult to hire for (in combination with good-enough context) include:

• People with management experience who are decent at it. Management capacity is a big bottleneck at many orgs.
• Experienced researchers, especially those with research taste —people who can come up with their own ideas and prioritise them, not just execute someone else's research vision.
• Founders who can take a project from 0 to 1.
• Competent generalists who can get a lot of shit done —the sort of employee you'd find in an early startup, someone who can solve problems without much hand-holding. This is downstream of the management bottleneck: organisations don't have enough managers, so they want 1) good managers, and 2) people who can do well without good managers.
• Communications specialists. This includes skilled writers who can make technical topics accessible, but also, more generally, people with good comms intuitions who can think strategically. People who know not just how to communicate clearly but when, where, how and why information should be shared.
• “Amplifiers/enablers” (particularly in the AI space) —this is a made-up category I use to refer to people who are happy to amplify others' impact rather than being the primary actors themselves. This includes program and research managers, operations roles.. and 80k advisors!

If you're decent-to-good at any of these skills —even if you don’t consider yourself exceptional— consider applying for relevant jobs. 

If you have significant experience in these areas and you're applying but not coming close to getting an offer, consider that perhaps you need to improve your cause area knowledge.

If you don't have experience in these areas but could get it relatively quickly, consider focusing on gaining that experience rather than finding an impactful job as soon as possible.

Be Careful About Deferring to Others

Beware of deferring too much on what counts as an impactful job. You've probably heard that the EA community defers too much and not enough people form their own opinions. This is such a strong meme now that I worry people are deferring to others about how much we defer…

But I do think it's true, particularly when it comes to impact, where complicating factors like status games get in the way of clarifying your own beliefs.

Common misconceptions include:

1. Believing an impactful job is the same as a job in an EA organisation. Most impact in the world happens outside EA organisations. And some roles in EA organisations aren't always counterfactually impactful (particularly if there are many competent candidates close in skills).
2. Believing all "impactful" jobs are on the 80k Job Board. The 80k Job Board is a great resource that I regularly recommend, but it has limitations:
   • Many job opportunities aren't advertised publicly. It is common to run closed hiring rounds where hiring managers pool a small number of candidates through referrals in their network
   • Sometimes, organisations aren't actively looking to fill a role but would hire the right person if approached
   • Many potentially impactful and usually neglected opportunities aren't "normal" jobs you can just apply for: founding an organisation, coming up with and executing a new project idea, working as an independent researcher or investigative journalist…

I’m sure many of you know you can apply for grants to pursue these non-traditional paths, but there's still a strong default to favour "normal" job openings. This default makes sense, but it's perhaps too overpowering.

Also worth considering: roles that may not appear impactful but could help you make a lateral move or get promoted to something more impactful relatively quickly. You should assess this on a case-by-case basis, but  some policy roles, technical roles, and “EA” jobs can be very meritocratic: if you do well, you can progress quickly or move elsewhere with good recommendations. Getting your foot in the door helps you gain experience, cultivate connections, and build a reputation. Sometimes, you're better off accepting a job that's not ideal but will make it easier to transition soon, rather than spending months applying for great jobs without success.

Multiple Bets Are Required to Solve a Problem

Making progress on a problem requires making many bets. We have heuristics about what might be important, but we don't know for sure. We need to cover all our bases so that when the right opportunity arises, someone is there at the right time and place. Sometimes, too many people focus on the obvious, high-status things that everyone agrees are impactful, collectively neglecting high-risk, high-reward opportunities.

For example, if you care about AI governance, you might think US federal policy is the most important lever to influence the future (although US federal policy is looking more and more uncertain by the day, but let’s not get into that). But even if that were true, it wouldn't imply that all other areas are ~irrelevant and effort put on them should be ~zero:

• The California AI Bill SB 1047 didn't pass last year, but many (including me) believe it would have been good if it had. Having more talent working at the state level would have been helpful. Now there’s some uncertainty about what is most hopeless in the US, if it’s federal or state policy, but the answer to that is maybe not to pick one, but to try to do both
• A few years ago, there weren’t many people with a good understanding of AI risks and a scope-sensitive mindset trying to shape the EU AI Act. A handful of individuals managed to make a big difference, plus build good career capital, in part because it was a relatively neglected space with little competition (the EU AI policy talent pipeline has improved quite a lot since, but it could still absorb much more talent).
• And individual countries can matter too: Spain was randomly more important than you’d have expected during the negotiation of the EU AI Act because they held the rotating presidency of the Council of the EU at a crucial time. France recently organised an AI Summit (which was somewhat disappointing). We need informed people ready to step up when these regional opportunities arise.
• In my opinion, we need international agreements to regulate AI, which requires knowledgeable experts pushing for AI governance from Latin America, Africa, and Asia—not just the US and Europe. Any political leader, in any country, taking AI GCRs seriously would be a big deal! 

I know it can feel demotivating to work on something when you don't know if it will make a difference. But we're all operating under uncertainty, and your chances of making a difference are higher with a relatively neglected approach. Think about expected outcomes, not "guaranteed" ones (nothing is guaranteed).

During COVID, hundreds of efforts sought to research and develop vaccines, hoping one would work. Only a few were eventually successful. But if every researcher had decided not to work on this because they couldn't know if theirs would be the one true vaccine, we would never have found one.

An important clarification: I'm not suggesting you work on whatever passion project you want because there's some chance it will be impactful. As an advisor, one common mistake I see people making is not prioritising enough or thinking enough about the very best opportunities. So I'm not suggesting you work on something neglected for the sake of doing something nobody else is doing. The point is that some people focus too much on what seems most important because it’s easy to have visibility into that, without thinking about what is neglected and tractable relative to its importance (and a good fit for their background and skills!). You need to consider how all these factors work together.

Be Prepared to Pivot

Be prepared to pivot when the opportunity arrives. This is particularly applicable to people worried about AI and short AI timelines, like myself, so I’m going to say a few words about that. 

If you think AI might dramatically change (or potentially end) the world in the next few years, it's natural to feel urgency and want to contribute sooner rather than later because time is running out. I personally share and endorse this urgency (to an extent). 

But it can put you in a counterproductive mindset where you're so focused on having an impact now that you're not thinking about opportunities to have a critical impact in 1-2 years.

For example, there was a time when AI wasn't a hot topic outside of EA. Back then, there were basically no AI policy jobs because AI policy wasn't really a thing. Some people built expertise in policy focused on unrelated areas, while staying in the AI loop, independently reading about AI and making connections. Then, when GPT-4 came out, policymakers worldwide scrambled to understand the technology and how to respond. This created an opportunity for those people with policy experience and basic AI knowledge to have a disproportionate influence.

I think this could be true for technical AI safety roles too. The field has grown significantly —there are many more roles now than five years ago—, but AI safety roles remain a tiny minority compared to capabilities roles, and the field will likely continue to grow with more openings in the future. If you have the opportunity to build useful expertise in an area like cybersecurity, or hone your management skills (even in a context that’s not optimising for immediate impact) you might be better off doing that and waiting for an opening to put that valuable career capital into use. 

I certainly believe with AI we're likely to see more windows of opportunity to act. These could be triggered by capability jumps that make governments take things more seriously, or by warning shots. What's important is that you have your own model of what game-changing events we might see in the near future, and how to prepare to be maximally useful when they happen. Preparation could mean building skills, knowledge, and connections, but also savings if money is a concern, so you can take a pay cut or move to a more expensive city if the situation requires it.

Practical Job Hunting and Career Advice

Impact Requires Unpaid "Work" (Unfortunate but True)

You need to internalise that having an impact almost always involves doing "work" no one is asking or paying you to do. This means educating yourself, reading, chatting with people with similar interests, learning from them, and building skills.

In a sense, this is always trivially true because if you want to have an impact, you need a model of what impact means and how to get there—your theory of change. This isn't something schools ask you to do; it's something you do for yourself because you want to and you care.

Once you have a theory of what problem you're trying to solve and how, you need to become someone who can contribute. You're unlikely to achieve this by relying solely on formal education or jobs. There isn't a structured path that you follow step by step to find impact at the end. This differs from most other paths, where, for example, to become a doctor, you go to medical school, get good grades, do placements, and if you do well, at the end of the process, voilà, you become a doctor! I wish we had a straightforward process like this for impactful careers, but for the most part, we don't.

Credentials for their own sake won't help much. If there's work you think is valuable and you could do well, start testing that as soon as possible —don't wait for an internship or job opportunity as an excuse.

• If you want to be a writer, write today. Start a Substack, write on an online forum, pitch articles to magazines.
• If you want to work on governance, start reading, thinking, and summarising your thoughts on promising interventions, where you agree and where you disagree with others, and why. Do not passively absorb information  —actively engage with it. For most fellowships and internships, having knowledge and opinions is a requirement to be a competitive candidate.
• If you're interested in technical AI safety, pick a small problem to work on. Try to replicate a paper. Write your thoughts on how promising different research agendas seem.
• If you wish something existed (a reading group in your city, an online community on a certain topic, an app that solves a problem you're frustrated about, a new research agenda), consider making it happen yourself.

Now, I completely understand and feel that it's difficult to get things done when you're relying solely on your own drive; it's easy to lose motivation. That's certainly the case for me! Creating accountability mechanisms with friends, coworkers, or productivity coaches can help.

If you're struggling to get things done, direct your energy toward setting up a good accountability mechanism, like a co-working group or just one friend you can convince to do pomodoros with.

Maximize Your Luck Surface Area

Go to events, speak with people, become an active participant in a community. This is the one thing that has probably helped me the most in my career. People with whom I've had a 20-minute conversation at an EAG have recommended me for jobs.

If you can't go to events as often as you'd like, reach out to those doing what you'd like to do. If you find an article you wish you had written or research you wish you had published, email the person responsible with a compliment and ask for a 30-minute call to ask follow-up questions.

Also, give yourself the opportunity to be surprised. Sometimes, you should go to events or meet with people when you're not entirely sure how or if they'll be helpful. The upside is limited when you already know how a conversation will be relevant. You don't know what you don't know.

Say "Yes" to Things

Collaborate part-time on projects. Volunteer.

Offer your help, and make it concrete. There's this common piece of advice for friends and relatives of people going through a rough time due to illness, mental health issues, or other life difficulties: If you tell them "let me know how I can help," they may not let you know, because you're giving them work. It's work figuring out what they need, whether you can provide it, and reaching out to ask for it. What you need to do is make it easy for them: "Would it be helpful if I bought groceries for you?" or knock on their door and ask, "Do you want to go for a walk?"

The same principle applies to employers and collaborators. Offer concrete help, identify the problems they face, and tell them, "I can help you with this."

Reframe Job Applications as a Communication Protocol, Not an Evaluation

Train yourself to reframe job applications as communication protocols, not evaluations. Hiring managers aren't judging if you're a talented person; they're trying to figure out if you can help them. And you're trying to figure out if you can help them —and if you want to.

One of the most common mistakes I see job candidates make is thinking they need to convince the hiring manager they're great –probably because this mindset of “this organisation is going to judge me, so I need prove my worth”. Organisations don't care about your being great per se; there are lots of great people they don't want to hire. Organisations have problems and needs, and they want to know if you’ll be able to help them.

From your perspective, a job application feels like it’s about you. But from the organsiation’s perspective, it’s about them. In your application, show them that 1) you understand their needs (what they’re trying to achieve, and where the specific role you’re applying for fits into the picture) and that 2) you can help, because you have relevant experience or skill set to address those needs. This could be from a different job title, through internships, volunteering, side projects... Anything! Offer evidence that you can do what they need. Which also requires spending some time to understand what they need in the first place! 

If you don't get a job, that's information! Could be that: 

• You may not understand the organisation's needs as well as you thought. This usually rules you out early in recruitment processes. If this happens, take it as a prompt to learn more about the space or cause area.
• You may not be able to help them as much as you thought. This usually rules you out later in the process when work tests are involved. If you don't pass a work test, take that as a prompt to reflect on why, and upskill.

This is why you should see applying for jobs as a useful process in itself. It's a skill you can improve at, and it's a mechanism to gather information on your strengths and weaknesses. Most people don't apply for enough jobs.

And I'll reiterate: try to get yourself in a position where employers reach out to suggest you apply for jobs, because then you don't need to decide whether it's worth applying —someone else has made that decision for you. This is an underappreciated reason why speaking with an 80k advisor can be useful. 

Final Thoughts

I've focused on jobs that could be directly impactful, but there are other considerations:

• Having a job you're good at and donating part of your income continues to be a really solid way to have an impact. In some cases, it will be the best way to contribute —not just a Plan B, but genuinely the best thing you can do.
• Remember there are ways to have an impact outside your professional life. If for whatever reason you have limitations that make it difficult to pursue certain jobs, there's still a lot you can do, like supporting others and communicating ideas outside of your work.

Best of luck with your job search!