jamesbrady

Elicit | San Francisco + remote (US timezones) | https://elicit.com/careers

Elicit is building AI for science. We scaled to >$7MM annual revenue and 450k MAU fast. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.

Reasons you might want to join Elicit:

1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (https://ought.org/) which was an applied research lab.

2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. https://ought.org/updates/2022-04-06-process for a primer.

3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.

Reasons you might not want to join Elicit:

1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.

2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.

A selection of open roles—more on the site:

• ML engineer: https://elicit.com/careers/913a03d5-bd26-4c64-8346-21a029f34...

• Founding data engineer: https://elicit.com/careers/8cc10be7-30ca-469a-a567-dfcdb271f...

• Senior software engineer: https://elicit.com/careers/aa99e2e9-5b15-4cd3-ac9d-9c9177ff6...

Elicit | San Francisco + remote (US timezones) | https://elicit.com/careers

Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We scaled to >$6MM annual revenue and 450k MAU fast. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.

Reasons you might want to join Elicit:

1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (https://ought.org/) which was an applied research lab.

2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. https://ought.org/updates/2022-04-06-process for a primer.

3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.

Reasons you might not want to join Elicit:

1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.

2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.

A selection of open roles—more on the site:

• Machine learning engineer: https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...

• Our first data engineer: https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...

• Front-end engineer: https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...

Elicit | San Francisco + remote (US timezones) | https://elicit.com/careers

Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We scaled to >$6MM annual revenue and 450k MAU fast. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.

Reasons you might want to join Elicit:

1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (https://ought.org/) which was an applied research lab.

2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. https://ought.org/updates/2022-04-06-process for a primer.

3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.

Reasons you might not want to join Elicit:

1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.

2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.

A selection of open roles—more on the site:

• Machine learning engineer: https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...

• Our first data engineer: https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...

• Front-end engineer: https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...

Elicit | San Francisco + remote (US timezones) | https://elicit.com/careers

Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We scaled to >$5MM annual revenue and 450k MAU fast. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.

Reasons you might want to join Elicit:

1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (https://ought.org/) which was an applied research lab.

2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. https://ought.org/updates/2022-04-06-process for a primer.

3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.

Reasons you might not want to join Elicit:

1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.

2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.

A selection of open roles—more on the site:

• Front-end engineer: https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...

• Machine learning engineer: https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...

• Data engineer: https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...

Elicit | San Francisco + remote (US timezones) | https://elicit.com/careers

Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We've scaled to >$3MM annual revenue and 400k MAU with our small team. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.

Reasons you might want to join Elicit:

1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (https://ought.org/) which was more of a research lab.

2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. https://ought.org/updates/2022-04-06-process for a primer.

3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.

Reasons you might not want to join Elicit:

1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.

2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.

3. We just raised a Series A, so if you want existential dread (and concomitant founding-team-upside), Elicit is not for you.

A selection of open roles—more on the site:

• Front-end engineer: https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...

• Machine learning engineer: https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...

• Data engineer: https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...

Elicit | San Francisco + remote (US timezones) | https://elicit.com/careers

Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We've scaled to >$3MM annual revenue and 400k MAU with our small team. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.

Reasons you might want to join Elicit:

1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (https://ought.org/) which was more of a research lab.

2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. https://ought.org/updates/2022-04-06-process for a primer.

3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.

Reasons you might not want to join Elicit:

1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.

2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.

A selection of open roles—more on the site:

• Front-end engineer: https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...

• Machine learning engineer: https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...

• Data engineer: https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...

Elicit | San Francisco + remote (US timezones) | https://elicit.com/careers

Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We've scaled to >$3MM annual revenue and 400k MAU with our small team. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.

Reasons you might want to join Elicit:

1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (https://ought.org/) which was more of a research lab.

2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. https://ought.org/updates/2022-04-06-process for a primer.

3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.

Reasons you might not want to join Elicit:

1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.

2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.

A selection of open roles—more on the site:

• Front-end engineer: https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...

• Machine learning engineer: https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...

• Data engineer: https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...

Elicit | San Francisco + remote (US timezones) | https://elicit.com/careers

Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We've scaled to >$2MM annual revenue and 400k MAU with our small team. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.

Front-end engineer: https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...

Machine learning engineer: https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...

Data engineer: https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...

Elicit (https://elicit.com/careers) | Oakland, CA and Hybrid | Frontend, AI, and Full-stack software engineering roles.

Elicit is automating high-quality reasoning so that we can help the world make more breakthroughs in every domain: from climate change to the gut microbiome to longevity and economic policy.

We’ve scaled to over 200,000 monthly users purely by word of mouth and recently crossed $1.5MM in annual revenue, 7 months after launching subscriptions.

We’re now building out our software engineering team, and hiring across several technical roles.

If you'd like to know more about some of the work we're doing you could check: - A recent blog post current UX work: https://blog.elicit.com/living-documents-ai-ux/ - Me talking about our "AI engineer" role on a podcast a couple of weeks ago: https://www.latent.space/p/hiring

Check out the careers page linked above, or email me at james@elicit.com

Elician here!

Our main focus is a little different to SciSummary actually. We're focussed on understanding researchers broader workflows, and providing a research assistant (i.e. rather than a particular narrow tool for summarisation or search).

The workflows we're most excited about at the moment are literature and systematic reviews: we think we can make these orders of magnitude faster and higher quality.

Actually, it's not an LLM!

We do use LLMs, but the secret sauce is an approach we call Factored Cognition which we wrote about here: https://ought.org/research/factored-cognition

(Elicit the company and app was spun out from Ought the research lab).

We do joke internally about the homophone (in fact, IIRC we did a little joke on our CEO by rebranding for his birthday in 2022) but I'm sorry to report that we're all careful, ethical, and well-behaved people :(

Oh, my bad—I misunderstood, thanks for the clarification

Elician here: thank you for sharing our tool and for this praise!

We're glad you're enjoying it.

Elician here.

This is a good point! (Hopefully) obviously, if we knew a particular claim was fishy, we wouldn't make it in the app in the first place.

However, we do do a couple of things which go towards addressing your concern:

1. We can be more or less confident in the answers we're giving in the app, and if that confidence dips below a threshold we mark that particular cell in the results table with a red warning icon which encourages caution and user verification. This confidence level isn't perfectly calibrated, of course, but we are trying to engender a healthy, active, wariness in our users so that they don't take Elicit results as gospel. 2. We provide sources for all of the claims made in the app. You can see these by clicking on any cell in the results table. We encourage users to check—or at least spot-check—the results which they are updating on. This verification is generally much faster than doing the generation of the answer in the first place.

Elician here! Thanks for your comment.

I'm not sure I agree that those rule-of-thumb statistics are "arbitrary" or "fictional"… I guess it depends on what you mean by that. I can say that on our part they're a good faith attempt to help users calibrate how best to use the tool, using evaluations of Elicit based on real usage.

Definitely accept that the tool can work better or worse depending on your domain or workflow though!

One way we do try to distinguish ourselves from vanilla LLMs is that we provide sources for all of the claims made. I mention this because we hope our users can approach the falsification process you mention for Google. We want to show people where particular claims come from such that we earn their trust.

Walking citation trails and verifying transitive claims is something we've talked about but need more people to implement! (https://elicit.com/careers)

Elician here!

Accuracy and supportedness of the claims made in Elicit are two of the most central things we focus on—it's a shame it didn't work as well as we'd like in this case.

I'd appreciate knowing more about the specifics so we can understand and improve

Elician here.

The two main problems we're addressing now are:

1. Finding papers / claims / data across an academic literature which is ballooning in size.

2. Using these raw materials to to answer questions in a reliable manner.

#2 is where the bulk of the tricky ML work is, and where vanilla language models often fall short because of limited context windows and hallucination.

We're also working to expand Elicit to help academics with other parts of their research, like surfacing critiques, suggesting related prior work, brainstorming related research questions, identifying risks of bias, …

This is not true.

Agree, way too long!

I gave that much of a heads-up of my leaving (not a formal resignation, as it happens) for a few reasons including:

1. It's a C-level replacement, which takes _months_

2. I cared deeply (and still do) about my team and the company mission, and wanted them to succeed: "the more notice the better" seemed right to me (it wasn't)

3. My career plans were to move into a new field which would require various courses, lots of reading, conferences, etc. – although I knew I was leaving, I hadn't even started speaking to potential next employers or investors when I told my CEO I was leaving

Putting all of that together is how I ended up with the 10 month blunder, but the post is less against 10 month notice periods, and more for matching the notice period to your handover, whatever that would be.

I feel your pain on working out those 5 weeks you mention! Awkward for everyone involved.

Thankfully, it wasn't a train wreck!

But yes, there were certainly some things I would have done differently.

For me, I tend to learn more from my or someone else's mistakes than from when something went swimmingly – the post is an attempt to capture some of those.

Your last paragraph strawmans me… but yes, two of my points are related to what you mention. Don't sabotage your team and don't give too much notice – I've seen several people do one or the other of these.

To be honest I think we agree about more than you make it seem!

What you did at your last job sounds exactly right to me – you were giving good feedback about what you needed, and when it wasn't heeded you left. Your manager didn't (or wasn't able to) hold up their side of that conversation and you moved to the "you should have left anyway" box. Doesn't sound fun, but we're in agreement on the steps there.

The 10 months thing seems to have tripped a few people up… I might try to rephrase that? Note that the section is titled "Avoid giving too much notice" and lists the reasons that 10 months was way too long. The advice is to match the notice period to the handover period, along with giving your manager ongoing feedback (as you did) if you're heading in this direction. Interestingly, one other commenter did a 12 (!) month notice period and it seemed to work in that situation.

Maybe that is where we've crossed wires on your last point too? I'm specifically not advocating for super-long notice periods, but as a hiring manager I would always respect a candidate who wanted to do e.g. 6 weeks instead of 2 if there was a good reason. For more senior roles, two weeks is not standard, nor is it internationally.

I think we might have crossed wires here…

The section you're referring to is titled "Avoid giving too much notice " and basically gives five reasons for why 10 months was a harmfully-long notice period.

The advice is to give much less notice than that.

I guess the source of the confusion is that you didn't specify what the sabotage actually is

Ah ha, I see – yes I had in mind actively doing stuff to undermine the team after the person has left (on the benign end, deliberately doing a crappy job of documentation; on the malignant end, stuff like data destruction). I will make this clearer in the post, thanks!

- The notion that you should have loyalty to the team beyond your last paycheck.

It made me sad to read this. You don't have to be loyal to your team, of course, but your framing makes me think that you've never been in a position where you've wanted to be loyal to your team, long-term.

It should read: "How Your Boss Would Like You to Quit"

Touché, yes there's some truth to this. Many of the points in the post were reflections on people leaving my team as I have an order of magnitude more examples of that than me leaving a team. If you're implying there's some kind of zero-sum thing going on here though, I disagree: it can be good for both you and your boss.

I wonder what their world view is…

I don't quite follow this paragraph, could you rephrase?

Leaving the team at short notice is unethical?

Ah, no – this is in the "Don’t sabotage, it’s a dick move" section: it's sabotage which is unethical.

But if you know that the current company cannot pay like the other company you are interviewing for, or that in this new company the future salary outlook is much better, why bother asking if they can pay you more?

That's situation-dependent: there's often stuff a company can do if you're highly cash-motivated (promotions, other title changes, bonuses, …) to retain you.

On the other hand, often there's not! Especially in older / larger orgs or ones with unconventionally rigid payscales. Even in that situation, I'd generally stand by the recommendation: there's a chance they will find a way to accommodate your needs, and if not you have cemented a strong relationship with your manager and above that will benefit you way into the future.

A few other commenters have mentioned the benefits they have realised by leaving gracefully. IMO this expected upside outweighs the short-term risk, at least in the sorts of roles the post is (now) aimed at.

Was this a tech company?

I just tightened up the intro explaining whom this post is meant for, and assumed that high-demand industries would be pretty immune to this kind of insanity

I like this phrasing and will steal it henceforth.

OP here, yeah I mention the 10 months thing as an example of what not to do – the actual handover of tasks I did in maybe 1–2 months, and that was at a leisurely pace TBH. The first ~9 months we were in a slightly weird limbo unfortunately – that's what I was flagging up as an anti-pattern there.

the assumption is getting fired instantly from a toxic organization is probably a good thing

That's correct – although I definitely take on board some of the other commenters' feedback about things like visas tied to roles. I'm going to add some clarification to make it clear this really relates to employees in good standing in tech companies (obviously a tiny minority of the populace and one of the few groups who have the guarantee you mention).

Yes! Good point: the manager needs to create the forum for those conversations to happen but the report still needs to take it on themselves to have that sometimes-awkward conversation.