My company is hiring. We've mostly given up on traditional application pipelines, there's too much AI spam and lying on resumes. Robin Hanson says that hiring is the killer app of prediction markets, so let's try it?
I work for a growth-stage fintech company innovating new ways to prevent fraud. We're based in New York, though most of our employees are fully remote. Positions are flexible; we're looking to hire people with a technical bent who are generally competent and excited to solve problems. Most of our positions are for software engineering, but we're also hiring at least one operations person who doesn't need to know how to code. (We also want some salespeople, but I don't expect to find many of those on Manifold.)
In general if you're on Manifold you'd probably enjoy working with us. It's a prosocial job, helping harden society the predicted coming wave of agentic hacks and social engineering. You can mostly choose your own tasks as long as they clearly make our product better, and you'll have significant input into the long-term direction of the company.
I list some some desirable qualities below; We don't expect anyone to have everything on the list, but you should have more than one.
A generally sane approach to AI. Far too many engineers nowadays are either A) refusing to use AI coding agents altogether, leading to low productivity and unnoticed bugs, or B) vibecoding everything and shipping barely-working unmaintainable slop.
Security mindset. This is distinct from general software engineering; I have met many otherwise-decent programmers who are garbage at security. The code they write works for normal users, but crumbles as soon as an adversary starts looking for ways to exploit it.
People and project management. For example, an engineering manager who has the technical and organizational skills to estimate how complex each task will be, decide on the most efficient order in which to do them, give feedback on code quality/architecture, etc. while also having the people skills necessary to deliver feedback whose gist is "this 4-hour task has taken you two weeks" and "your code is trash" in a way that is more suitable to a traditional professional environment.
Consciousnesses and scrupulousness. Our clients are trusting us with billions of dollars and very private information about their companies. We need people who can be trusted to handle this information responsibly, both on a personal level and in terms of writing high quality code that isn't going to break.
Comfort dealing with compliance paperwork, procurement, internet research, meetings with nontechnical users of our software, repetitive maintenance tasks, and other work typical of business operations.
A solid knowledge of modern SaaS infrastructure: GCP, Cloudflare, Postgres, Celery, etc.
And by far the most important: INTEREST IN THE WORK. We're not a crazy SF grindset company, we have normal working hours and generous time off. But we are not interested in people who are on paper qualified, but once hired write like 50 lines of code a day and can't be bothered to take any initiative whatsoever. The ideal candidate is altruistically-minded, agentic, and intrinsically passionate about what we're working towards.
I am intentionally not naming the company here, but feel free to ask me if you have further questions. Salary is negotiable depending on what you bring to the table; we're open to people at any level of prior experience. We're potentially open to part-time employment, but nothing below ~20 hours a week, and for most positions we'd prefer full-time. We strongly prefer Americans, but have hired non-Americans in the past if they're a good enough fit.
Each answer here is an independent conditional market on the question: if we interview you, will I be happy with your performance in 9 months? (If we interview you but do not hire you, resolves NO.) Note that these will resolve based on my personal judgement of the quality and quantity of your output, not any sort of formal performance review. By adding yourself as an answer here, you consent to the risk that I may resolve your answer NO, even if your superiors and other coworkers are happy with your work. I have a relatively high bar, and would probably resolve to NO for the median software engineer in the United States. (If you don't make it 9 months, resolves based on my best guess as to what the answer would have been. Probably NO if you were let go, probably YES if you were poached and your work seemed high-quality until that point, etc.)
I will consider reaching out to anyone in the answers here, strongly favoring those at a higher probability. I will also take into account your history on Manifold, whether I already know you or your work, whether you have previously demonstrated competence or interest in a relevant field in a difficult-to-fake-way. Traditional credentials (e.g. a formal degree, past positions) are also helpful but not required. Feel free to provide me with any such information.
(If you wish to keep your Manifold identity separate from your professional, you may PM me to consider you without adding your name to the market. You had better have some other credible signals of your competence in that case.)
If you recommend someone we end up hiring who wouldn't otherwise have seen this market I'll give you an M$10,000 bounty.
I won't bet in this market.
1,000Providing some additional context:
My primary goal here is to hire some of the people that I've known from Manifold for a long time, having already credibly demonstrated their intelligence, resourcefulness, and reliability to me. I've reached out to a few directly, but I'm sure I'm forgetting some, and I'm hoping they see this market.
People that are newer to the site, or whom I have not previously interacted with, are welcome to throw their name in, but you shouldn't expect anything unless you provide me with strong evidence that you are worth my time and the reputational risk of recommending you. Your answer being high-probability is relevant evidence, but I am well-aware of all the ways that could be inaccurate/manipulated, so additional facts would also help your case. Understanding what kinds of facts are and aren't credible evidence to a recruiter will also benefit you, here.
@IsaacKing hmm, so you're saying I shouldn't treat the 1k to add an answer as an application fee? Too late now. At least you can see I'm biased towards action ;)
I'm not actually looking for work rn, but give it 6 months and I could be convinced. Given the close date of this market, that seems fine. I'm based in the UK and not willing to relocate, but happy to work remote.
The biggest point on here I resonate with is the security mindset. I like to write code that works 100% of the time, which is frustrating because in most instances <100% is optimal. I'm also good at spotting (and exploiting) the unhandled 0.001% in others' code. I'm particularly interested in fingerprinting, which I know is a common element of anti-fraud technologies.
My biggest weakness on these criteria is project management - The closest I've been to management is telling Claude what to do. When it comes to people though, I think I have reasonable tactfulness.
I'm most interested by the prospect of "you can mostly choose your own tasks as long as they clearly make our product better"
I have a history of being hired via internet fora.
@retr0id I've seen what you can do in FOLDED. I think you're one of the frontrunners here.
@bens not at all, it's the right move. Isaac might want higher numbers as indications, but the odds are strongly against everyone by the natures of hiring in general and the market structure itself
(I do not think Isaac’s company will hire 3 Manifolders on expectation but what do I know)
directionally agree but IIUC this is a conditional market (N/A if they don't interview), so it's hard to interpret the sum of the probabilities here? e.g., if they typically hire ~25% of candidates they interview, and they plan to interview a few candidates, and the market can't really tell which candidates are any good, then you would have many options near 25% and while they would sum up to a large amount that doesn't tell you much about the total number hired in expectation.
or i'm misunderstanding what "independent conditional market" means, and candidates they don't interview will resolve NO, but that wouldn't normally be "conditional"
@Ziddletwix oh I guess I kind of assumed most ppl on this list were inviting an interview? But maybe Isaac has a higher bar for doing an interview? But also, there’s a further criterion of him being happy with their performance after 9 months.
@Ziddletwix ah yes, I forgot abt the conditional part. we don't know how many folks will be interviewed. my read, which may be off, is that Isaac is seeking a limited number to interview bc he wants the referrals to stick the landing. the risk is still high for NO for anything that isn't NA but if I'm right YES may be a more available result than we think
This market got more traction than I expected! Very exciting. Still, gonna throw my name up and pitch. Might as well, eh? Isaac's presentation of the role and answers to my questions were compelling. I'm particularly attracted to the prospect of autonomy in the face of diverse challenges.
In game industry terms, I often call myself a "systems designer," which has a different meaning than it does elsewhere in software; it's less code architecture and more about gameplay and mechanics design. My mindset about systems goes beyond game development, however. I see the world as a layering of systems: economies, ecologies, cultures, communities, languages, resource flows, incentives and disincentives. My way of understanding anything is usually to look at the little knobs that make up its intermingling parts, whether the context is physics, biology, economics, or even something softer like social interactions.
I've got a ton of varied skills from 15 years in games + a couple careers before and concurrent: project, production, and other management, programming (off the top of my head, I've worked in varying degrees with C#, gdscript, Java, JS, TS, python, C++, and old school html. I'm sure that's not all of them. Recently I started playing with an assembly language), devops (mostly CI/CD), technical architecture, a myriad of design disciplines, as well as some BI, art, writing, and a ton of community and culture-building stuff.
Much of that was self-taught or learned in one trial by fire or another. I'm an irrepressible autodidact driven by two motivations in general: a constant hunger for new challenge and a desire to autonomously improve the experiences of those whose lives I touch. In business words, I seek efficiency and self-improvement, but I do that so I can ease the burdens of coworkers and customers.
A VP from Microsoft made perhaps the best case for hiring a gamedev outside the context of games.
I didn't want to make this much longer, but I think for the sake of this market it's fair to address the desirable qualities Isaac is asking for, including where I might fall short:
A generally sane approach to AI.
I'm bearish on AI. I do dabble with it as a way of keeping tabs on its capabilities and testing how it might improve my own work, but I'm certainly not a fanatic about it, so I lean more toward luddite than vibecoder. Tangent: I believe LLMs are structurally flawed and fundamentally unable to grow meaningfully to the ASI people are slathering over because of their neural network foundation; genetic algorithms, while a rougher fit for early development, have significantly more potential.
Security mindset.
I have done some security work, but it has been in the context of games. We have a great many technically-savvy bad actors in that space (hackers, cheaters, scammers, botnets, child predators) who I've had to build designs and architectures to thwart, but ultimately our tech stacks and products are built with an eye to UX, not robust testing suites and critical financial infra.
People and project management.
I def believe so, but soft skills can be hard to prove to oneself and to others. I've spent a lifetime studying people, motivations, and leadership, and I've dedicated some portions of my professional life to the specific skills involved in project management.
Consciousnesses and scrupulousness.
afaik I'm generally considered a reliable and trustworthy person. I tend toward being a perfectionist in my work, which comes with as many downsides as upsides but sounds like a pro in this case.
Comfort dealing with compliance paperwork, procurement, internet research, meetings with nontechnical users of our software, repetitive maintenance tasks, and other work typical of business operations.
Except research and interfacing with non-technical users, those aren't among my favorite things. 4/10 comfort.
A solid knowledge of modern SaaS infrastructure: GCP, Cloudflare, Postgres, Celery, etc.
Most GaaS use different stacks than SaaS. I have some knowledge of the principles but not those specific tools.
And by far the most important: INTEREST IN THE WORK. ... The ideal candidate is altruistically-minded, agentic, and intrinsically passionate about information security, protecting people from fraud, and generally making a high quality product.
I'm for sure altruistically-minded, agentic, and passionate about making high-quality products. The other points are more or less yet-to-be-tested; a lot of my motivation and interest in a topic inevitably derives from the challenge and sense of purpose it provides.
@nsokolsky The problem with providing a numerical range is that A) it depends heavily on the candidate's skills and BATNA, making the real range too wide to be particularly useful, and B) people tend to anchor on the upper end of any range given to them, meaning that announcing the true range can lessen the chance of successfully making a deal later on.
@IsaacKing let me rephrase then: assuming this person was good enough to be hired by DeepMind, would you be able to offer them at least 50% of Google’s likely TC in liquid assets?