[State of Research Funding] Beyond NSF, Slingshots, Open Frontiers — Andy Konwinski, Laude Institute

Light in space, light 2025. Break ups. Light in space, light 2025. We're here with Andy Konwinski of Laude Institute. Welcome.

Yeah. Thank you so much.

I-it's weird to welcome you to your own launch show. We'll talk about that.

No, I, I am welcome here.

Yes.

I-- Everybody that-

Everybody

... invited is welcome here.

So your intro is very, very long, but you're a co-founder of Databricks and Perplexity, and, and, and now I think, um, people are less familiar with the Laude Institute and what you're-

Brand new

... what you're doing. It's brand new. We covered your Terminal Bench events recently.

Yeah. So fun.

That was little bit of a coming out party, but maybe just for folks who are out of the loop, what is the Laude Institute?

Oh, good question. Laude itself, beyond just Laude Institute, is an organization that gives resources to researchers at the exact moment that they need them. So we say right resource, right researcher, right time, and there's a sort of dotted line in the life cycle of a researcher who makes a company, which is when you incorporate.

So after that, you wanna raise venture funding. So Laude has a venture arm that funds researchers and researchee technologists that aren't coming from PhDs or, or professorships to build companies, and that's actually a path that I am extremely passionate about with Databricks, Perplexity being companies founded by PhDs at Paradigm, at the heart of a paradigm, respectively big data for Databricks and public cloud, and then kind of search as GenAI came around for Perplexity.

So that's a, a part of Laude that I'm very passionate about, just helping researchers build the next Databricks, the next trillion dollar or n- so far hundred billion dollar company. And before that dotted line, researchers need, uh, in an unprecedented way, like an existential, existential way, more resources to participate in frontier AI research, uh, people doing research in the open.

And inspired by Databricks and my history, I want big breakthroughs to happen in the open from research, things that move humanity forward, like computing, you know, silicon or the personal computer, internet or big data with Databricks or, uh, the next paradigm of search with Perplexity.

So I want all that to be happening in the open, so I wanna fund the researchers who are doing open research just in a philanthropic motion. And across both of those, there's-- So, so there's Laude Ventures for the venture fund, Laude Institute, this nonprofit for the upstream stuff, and across both, this obsession with shipping, uh, in, in the case of the venture fund, shipping your code as product, in the case of the nonprofit, shipping your research as open source projects.

And then there's this ethos to bring meritocratic principles, which researchers love, into the whole thing as a governing philosophy, so for researchers and by researchers. So the venture funding, this fifty professors and PhDs, the top names like Jeff Dean and the top faculty at Berkeley and Stanford, and my co-founders of Databricks and Perplexity, who've invested in the fund and then give advice and make deal flow introductions.

And on the institute side, it's my money as, as sort of a successful technologist, plus the tech-- the money of other billionaires who have done well by their tech companies giving money into the nonprofit to do this no strings attached grant writing that ultimately becomes a funnel for recreating the Databricks motion.

You know, like make the breakthrough in the open, go become a billionaire tech founder. That translates your research insight into people-- products in people's hands. So that's Laude, Laude Ventures, Laude Institute, just kind of this one Laude vision.

Laude visions. Let-- I'll, I'll dive in. I wanna talk about Slingshots and as, as a funding-

Yeah

... vehicle. But you mentioned a lot, a bit of the, the sort of Databricks motion, and I think that's an interesting analogy because Databricks is very unusual as a founding story.

Hmm.

And I'm wondering if it's replicable. I thought it-

Why do you think it's unusual?

Uh, eight found- co-founders.

Seven, eight.

Yeah. So basically, it-- what was the model that you think really worked for Databricks, and does it transfer to the other Laude grantees who are not as big, who don't have the same-

No, I think OpenAI had quite a lot of co-founders. I think VMware had a decent number of early research team turned co-founding team. I think Snorkel AI is a unicorn, came out of, out of Stanford to have a pretty s- good size founding team.

Core research teams, when they gel at a university or outside at a research lab in an industry-

Yeah, they come as a unit

... with the years-

Yeah

... of, of deep scars from working together, knowing each other's ins and outs and team, like the, the dynamism and, like, the, the gestalt, like sum is greater than the parts of working together as a team, they're great bets.

So you kinda de-risked the founder divorce risk- ... with, uh, that's one of the most big problems that VCs have to watch out for by having built a breakthrough together. You've also pr-proven traction that you know how to come up with a disruptive idea.

You still have to prove that you can turn it into a product market fit. But I think that the founding team size is, is one thing that there is some precedent for. The path from... I cited a few examples of the path from research, just massively successful startup.

The Google co-founders, Larry and Sergey, obviously took research, NSF-funded research, by the way, and, uh, turned it into one of the most iconic companies in the history of humanity. I cited VMware. There's even examples, you know, more and more examples of researchers besides Perplexity coming out of Berkeley, like LMArena more recently has become a de facto standard for evaluation-

Models

... for all of the, the, the whole ecosystem, and that was a research p- turned startup product now, project now. So, uh, ten years ago, I would've agreed that it was... And when we founded Databricks, sure, it was a little bit more rare.

Now, I think it's becoming the gold standard, and the thing that, you know, all the VCs want in on. And being at, inside the heart of it all, Laude Institute sort of bridges those two and is obsessed with that path to impact that so many researchers wanna have.

So I would say that, um, I do believe going, you know, we are, we have reached the tipping point already of this being the most high leverage path from breakthrough to breakout company to world-changing. You know, hopefully you'll get to that Trillion dollar- ...

or beyond t- we'll g- some people will be at $10 trillion in value. Uh, so I think it's gonna be the new norm for... And, and it'll be great for the world to have this path, 'cause researchers are-- they're both amazing founders, and they're genuinely good people in that they tend to want...

They're very utilitarian. They want, they wanna bring all of humanity forward, very pragmatic and reasonable and evidence-based. So kind of, I think the world needs a little bit more of much of that right now too.

Yeah, amazing. Uh, let's talk about some of the research directions that you have funded through your Slingshot program. Um, so we, we already featured Terminal Bench. People can go look at that episode. Uh, Arena, we already mentioned. We also just did an episode.

Um, one thing that we haven't talked about as much is JePA, which-

Yeah, yeah

... y-y you have a couple of JePA, uh, co-authors in there.

That's right.

Um, maybe JePA and then any others that you care to pick up, but-

Yeah, I've been asking people what's been buzzing as I've been talking to people at NeurIPS this year, and this idea that the language models themselves are more like compute or GPU a generation ago, where what can we build at the layer above?

And in software systems, we've traditionally thought of VMware being a great example. You have the operating system and the underlying architecture. Now how do you make it a layer of abstraction above that empowers higher leverage decisions and choices and more applications that are more profound and powerful?

So I think that's neat to think that there's this layer above which is getting a lot of attention in the last year with context management, RAG being the most well-known kind of implementation of this. It's this layer above where we're managing what the, how the agents work and how they make decisions, how many memories they can keep tra- how and how many memories they keep track of.

That's all at this prompt management, and it's not just prompt anymore. It's like tool usage and, and memory curation and lots of innovation happening at that layer above the core stack, pre-training, post-training, distillation and, and now we're talking about one of the-- Dilara, the PhD student working on some of this, this prompt optimization, JePA-related stuff, talking about post-post-training, and that's tying over-

Ah, post-post-training. I like it. I like it.

Yeah.

Yeah, yeah, exactly.

So there, that's tying over to, um... There's actually a little bit of bleed over where this layer above now what we're talking about is compound systems or prompt optimization or in c- in, in context management is kinda getting a new level of buzz and a new level of, of enthusiasm and maybe a level up of the number of researchers working hard at that layer.

JePA, I would say, is the probably up there for or the most successful example of adoption so far of-

Over DSpy?

Well, it is kind of embedded and adjacent to it, so kinda like-

Yeah. I, I don't know what to make of it, you know? It's like DSpy has, has like a couple years of branding and-

Yeah, yeah, yeah, yeah

... has a brand.

And we're in-- DSpy, uh, the way I think of it is DSpy is something closer, more akin to a LangChain, so it's a, a framework for writing agent-type codes. You give it a task, and it'll, like com-- It, it, it takes natural language and compi-- uh, it actually takes code and compiles natural language, so it's like a reverse compiler.

That's DSpy. And the optimizers, which ca- kind of were born in the context of this DSpy project, they, uh, take a prompt and make it better. So that-

So JePA is a part of that-

JePA is like-

... optimizer

... flagship optimizer.

Yeah, yeah.

And it's using this evolutionary genetic technique, like a, a, a very old area of research reinvented by, uh, Lakshya, the main PhD student on JePA.

Yeah. I met him at the Terminal Bench.

And it's just like, really the way I zoom out and think about it is this layer above the core AI model stuff, and now it's broaching into doing model weight updates as well. And there's this big debate about is it in the context, is it RAG plus vector database plus whatever and sort of like lightweight, very few-- Like if I, if I tell you I had yogurt for breakfast this morning, you can remember that.

It's a very bespoke piece of information that I just told you. Uh, you didn't have to, like, go pre-train your brain weights for another round of like ten thousand GPU hours.

No.

You're able to-

It's an innate way, yeah

... way of remembering that. So that's kind of the parallel with this. And that also takes us, not to get too sprawling here, into continual learning, another buzz area that overlaps a ton with this layer. So really excited about that.

That higher layer co-compound systems is kinda a word that people are using to capture it. Continual learning factors really closely into it, context management, prompt optimization, JePA, better together. Those are some of the buzzwords floating-

Yeah

... in that, in that area.

A lot of what your source of alpha is, is coming from the Berkeley ecosystem. Uh, obviously you're not exclusively Berkeley. There's a lot of Stanford.

More and more Stanford.

More and more Stanford.

I'd say kinda equal partnerships.

Right. Is there a risk of that you're just doing only West Coast universities and-

No

... like, you know, the good ideas come from anywhere, right?

That's true.

So-

And we have massive focus to beyond Berkeley, Stanford.

Yeah.

I would say it's power law, big focus-

Which obviously like-

... on the background

... let's, let's be objective. Berkeley and Stanford have produced a lot, so that's fair.

Yeah, yeah. More than anything else.

Yeah.

But ta-- walk around here right now and you'll talk to Jackson Clark, second-year PhD from UIUC. I think second year. First-second year from UIUC in Illinois. Several UIUC teams here. He's running a thing called SRE Bench.

Wow.

Uh, we have teams from Stanford and MIT. Uh, sorry, not Stanford. CMU and MIT very well represented. Wisconsin PhD students around here, Caltech PhD students, lots of great projects. Our, of our Slingshots, I would say the majority are non-Berkeley, Stanford or something.

Maybe I don't... I need to double-check that number, but lots of great projects and a heavy push. One, one example is I started this PhD club focused on entrepreneurship at Berkeley when I was a PhD student in 2012 called Computer Science Grad Entrepreneurs.

It actually transformed in my brain eventually to this venture fund called Computer Science Grad Ventures, CSGV, and that was a proto-prototype that turned into Lawd eventually.

Ah.

So that's this lineage that-

So you've been doing this a while.

PhD. Yeah, yeah. 2012, PhD club. Now we've taken that. I've taken that, and I went to University of Washington, so starting with another West Coast school, and we started a club called Agent. Their, their CS department is in a building called Allen, uh, so it's Allen Graduate Entrepreneurs, Agent.

It's a PhD student club around entrepreneurship. Then we went to Wisconsin and started one called Research to Impact. There's another one at UIUC forming right now. There's another one at, one at Stanford called Saplings. So but a big focus on going to find the PhD students who are interested in shipping their research at CMU, UIUC, MIT, Wisconsin, Caltech, all the top 15 universities, Toronto, Waterloo- And, uh, McGill.

Yeah.

And then the, the, like, Vector and Mila. So North America, bringing Canada into the mix 'cause they've got also equally great universities up there.

Yeah.

Uh, there is a really big focus on bringing, creating a fat pipe of bandwidth to the root of where all the action... Like, you can't get around that the root is, of all of it's happening in Silicon Valley.

Yeah.

It's really opening the, the b- Like, flying them out here more, engaging them deeply in the program, giving them money.

And, uh, let's be honest, getting you as a mentor.

That's right.

That's right.

And, and I'm collecting other ... Braden-

Yeah

... who, who did Snorkel. He was a PhD of Chris Re at Stanford, who went on to do a unicorn company. He joined Laude. So assembling a partnership, like a venture partnership, but for PhDs and researchers turned founder, turned unicorn founder.

Yeah. I wanna get to Open Frontiers, but one more question about just the NSF, which I think you, uh, you mentioned once for, for about Databricks, but also I think you're trying to target the Laude grants as sort of NSF-level sort of prestige and, and-

Mm

... and, and impact. I think one thing I, maybe as like a spicy question is, well, what's broken about the NSF process are you trying to fix?

Uh, I don't think the NSF is broken. I love NSF.

Okay.

It has been probably the best investment the American pop- popular, uh, public has ever made. Thousands X return on their investments. I mentioned Google earlier, Databricks, all these researchers ca- came from NSF funding. Uh, and it's just been a paradigm generation.

So DARPA was important first. At, at some point, there's been these, these paradigm shifts in how open research has been funded. DARPA was a key one. NSF's been a key one, and now NSF is not, it's not big enough.

W- It was $1 billion a year for computer science, and that they're trying to cut that into half of that, but we need $10 to $100 billion to do frontier AI research. So, uh, it's not broken. Uh, they are trying to break it-

You need more

... but it is insufficient.

You need more NSF.

And, and I think in addition to the mechanisms NSF has used to dis- to boil the money, which are very effective, we have secret sauce in Silicon Valley of how startups find product market fit. We have, uh, really good pickers.

We have venture capitalists that, that have their own sort of evolutionary, uh, Hunger Games way of, like, finding who is great, who are really good at, at having intuition when they meet a person to pick the next winner.

That's a, that's a, a unique a- approach that NSF does not use. They do not make a little venture partnership-inspired group of research partners who go grant, do grant writing. You don't get equity back, but you do, uh, empower the right project.

If you, if you've got good pickers, you can actually, you can actually have much more effective deployment of a way sm- You can, you can actually deploy an order of magnitude less capital and have more impact with this approach, and it's very complimentary 'cause you can still have traditional NSF.

NSF is, is brings faculty, many of which are very near and dear friends, into a centralized organization that is in charge of deploying billions and billions of dollars of funding, has an insane track record, and we are gonna complement that with a much more Silicon Valley-inspired approach of a high velocity, turn the funding around really quickly, have very strong opinions about the type of research we fund and re- and be very narrow in the type of research.

NSF funds everything fro- like, all types of research, from biology to, you know, like, computer to, uh, and beyond, uh, English research and anthropology and sociology. Like I said, $1 billion to computer science, $10 billion total, so one tenth going computer science, and NSF has to manage all that.

Laude and Laude-like approaches were focused very tightly on high-impact computer science research.

Yeah.

Especially AI, but AI systems, you know, cryptol- cryptography and security and networking and architecture. Um, so, like, other areas besides core AI as well. So by being so laser-focused, we can go deeper. By bringing in p- researchers who have shipped, researchers who started companies, researchers who had found product market fit, we can actually identify projects that are more likely to become a Databricks or an Apache Spark or a Ray or an LMArena sooner in f- and, and with more confidence.

And so I think it's a very complimentary model-

It's-

... but nobody's ever done it before, and you needed someone to kinda step up and propose a pretty fundamentally different architecture or organization shape. Like, you know, hire people in a different way. It's a nonprofit, which is, you know, got its own challenges, but you want it to behave in a very lean, extremely high velocity way, like a startup.

So that's what we did, and it's working amazingly so far.

Congrats on, on everything. Final question. Uh, we have the Open Frontiers, uh, announcement.

Yeah.

And you're also doing the Laude Lounge here in NeurIPS. How are you get, able to get such amazing people? You've got Oriol Vinyals and Jeff Dean and, uh-

And so-

... François Chollet.

Yoshua Bengio.

Yeah.

And Dylan Patel.

That was a lo- long one, yeah.

But so fun.

Uh, so, so basically, like, uh, what is Open Frontiers? Where is this all going?

Great. Yeah. So Laude Lounge, basically NeurIPS has needed a VIP lounge. We have VVIPs, the gods of computer science and AI, as you know, obviously AI particularly, walking around the halls of that Giants conference hall.

Yeah.

It's actually hard to find a chair, much less a comfortable couch.

Yeah. You got meeting room 23A.

Oh, nice. It'd be good.

Uh.

Yeah. No, no, no, it's, it's not a VIP launch at all. It's just super-

Oh, it's-

Yeah, yeah. So the idea here is open 8 to 10, 8 AM to 10 PM. Identif- And it's not too hard to know who the big names are, the high-impact researchers, Jan Leike, François Chollet.

Yejin-

Lex Fridman.

Yejin, Dee- Dee's over there, and Amit Talwalkar, and we kinda know the circles already. So it just took a lot of emails and texts to like, "Hey, if you're gonna be at NeurIPS, across the street we're just gonna throw a VIP lounge Wednesday, Thursday, Friday.

Free food, Starlink Wi-Fi, open all day."

Chat with us for a little bit.

"Food's happening."

Yeah.

You wanna jump on a podcast, you wanna talk to some, so talk to Swyx. I mean, you're a big draw of people-

Wait

... coming out here too. Uh, walking through, people are like, "Oh, is that Swyx?"

Yeah.

Yeah, like you just walk through and you feel important, and you are important if you're hanging out here. So it's, it's working.

So it's like the MVP-

Very

... of Open Frontiers.

Okay. So that's, that's Laude Lounge.

Yeah.

Uh, Open Frontier

Is a project that I've been cooking on for a while in my brain, but really it's come together in the last three weeks really quickly. It's inspired by the core premise of open research that happened, that, that is symbolized by NeurIPS, right?

People publishing their ideas. And what's happened in the last year is Western open science and research discourse has no lo- has, has lost the number one spot to China. So, uh, we-- if you ask Stanford and Berkeley PhD students, which I have dozens of them and they're hanging around here, I can introduce you, where, where are the best papers, the most interesting papers about AI coming?

They'll say, "I read twice as many interesting papers by Chinese startups than I did by Americans this year."

'Cause, 'cause they just make an effort.

Yep. Moonshot, Kimi, DeepSeek, they're publishing really interesting stuff. They make an effort to talk about it. Whereas in the United States, since OpenAI closed their doors and stopped publishing, so did all the other labs.

Yeah.

And by and large, you go and you don't get to publish. You are actually working on the frontier at those labs, but you're not talking about it.

Yeah, yeah.

And so-

So you need to open the frontier.

The, so the, the openness is the key word. Um, OpenAI was called OpenAI, and they were open for a while, and now they're not. We need something that fills in that gap to be a champion and bring together all open researchers to put forward a unified front a-at the fr- so that we can operate at the frontier.

And so that's what the goal of this thing is, to bring together every single organization that is leading in open research. So Allen Institute, François will be around for Arc Prize Foundation. We have Berkeley, Stanford, SkyLab, Berkeley AI Research, BAIR.

We got, you know, the, like, Yejin's lab and Marin, the project of Percy Liang, who's kind of very senior at HAI and runs the Center for, uh-

Foundation models

... foundation models.

He coined the term.

Exactly, yeah, foundation models. So Per- and Percy was in the lab with... A lot of this is just my, my network because-

Yeah

... Percy and I were in the same lab with Matei and all the other Databricks founders back in 2008- ... back at Berkeley. So it wasn't hard to go find these people. Everyone I've approached said yes. The idea is like, let's team up, get together one day in San Francisco in the next five months and get the hundred most influential open researchers together for a conference and for meetings where we share our roadmaps and we talk about common goals amongst the entire ecosystem.

It's kind of surprising this hasn't happened yet, and the, the, the thing we're basically doing is making an open frontier lab-

Yeah

... effort, starting with a conference.

Yeah. And it's live streamed, so it'll actually open-

So that we actually-

You can actually see

... like embody the democratic principles that have gotten us to where we are today.

Yeah.

Get an order of magnitude more people in the world watching the breakthroughs that are happening in these labs. All the continual learning, prompt optimization, BLM, SGLang, breakthroughs in inference, breakthroughs in evaluations, Terminal Bench, and a lot of other-- I just heard for the first time about this benchmark called Impossible Bench.

It's a genius idea.

Yeah, we talked about it, uh-

Yeah

... with, uh, Jiayun.

Heard of it until, yeah, just-

Yeah.

So, um, that, that dissemination of what's actually happening will capture the world's attention and, in turn, the goal is to, with the world's attention, help this become, to gel into a, the, a well-enough funded and resourced collaboration team-up of the ecosystem that we can once again become number one at open frontier research- ...

uh, because that shouldn't-

It shouldn't be so hard.

It shouldn't.

Yeah, yeah, yeah.

You know, everybody I talked to is like, "Yeah, it shouldn't." And then I was like, "How about we just put this conference together?" They're like, "That's a good idea." Okay, well, I'm doing it.

Well, thanks for kicking it off. I'm excited to support it in any way I can, and I'm excited to see you next year.

You will-- I'm sure you'll be there, and I'm sure everybody will be like, "How did we get on with Swyx?"

Thank you so much.

Yeah, my pleasure.