Snorkel AI | Jumpstarting Data-Centric AI

Hi everyone. Welcome to Grey Matter, the podcast from Greylock, where we share stories from company builders and business leaders. I'm Heather Mack, head of editorial at Greylock. Today, Greylock General partner Sam Motamidi interviews Alex Ratner, who is the CEO and co founder of Snorkel. Sam and Alex discuss the rise of large language or foundation models in AI and the role Snorkel plays in helping enterprise organizations adapt and deploy these models. You can read a transcript of this interview on the content section of our website greylock.com blog. And if you aren't already a subscriber to Grey Matter, please sign up wherever you get your podcasts.

The latest addition to their platform, released in November, enables enterprises to put foundation models to use. Today we're going to talk about what that looks like in practice and I'm pleased to welcome Snorkel CEO and co founder Alex Ratner. Alex, thanks so much for joining me on Gray Matter.

Thanks so much for having me.

Sam, as you and I often talk about, AI is a very dynamic and fast moving field. And even since you launched Snorkel, we've seen a lot of change. Let's start by just putting everything in context. Where are we today in AI and ML and how do you characterize Snorkel's role in its adoption?

Yeah, it's indeed a fast moving space and it's exciting every day. So a lot of where we started and we'll get back into this is this shift that you talked about from from model centric to data centric AI development and I'll start there and it's still obviously where we think a lot of the core focus deserves to be at a high level. This is an idea that the pain points or the blockers to AI development and deployment or you could more optimistically say the actual areas where an AI developer can productively iterate used to be all around the models, picking out features, building custom architectures, you know, building bespoke infrastructure, all of that, what we call model centric development and data and training data. The data models learn from used to be kind of seen as a second class citizen. I call this sometimes the Kaggle era of machine learning where you know, a machine learning developer's journey started by downloading a data set that was nicely labeled and curated and then trying to train their model on that. So fast forward to today. A lot of the machine learning technologies and models have just left forward. We'll talk about even the recent progress over the last couple months around foundation models in a second and but really over the Last couple of years they become more powerful, more push button, more automated, more standardized, more commoditized to the point where, you know, state of the art model is a couple lines of Python code and an Internet connection to get it going if you have the data. So the trade off over the last bunch of years, certainly the last seven or eight that we've been working on this, this data centric AI movement out of Stanford and now the company, the game has really shifted towards the data and how you label and curate it to teach machine learning models. The reality in most enterprises that we work with, top 10 US banks and government agencies and healthcare systems, is that if you want to actually build a machine learning model for something, the balance of effort or time might look like a day to get the model and maybe person months to label the data to teach that model. So that is, you know, the trend that we started with that the field has been going under this kind of shift from model to data centric development. And it's still the key thing that enterprise is struggling with and that we address now. On top of that, there have been some really exciting developments recently around what are often called large language models or I'll call them foundation models in this chat, partially out of loyalty to my co founder Chris Some, you know very well and he's one of the co founders the Stanford Foundation Model center. And partly because I think it's actually an appropriate name, these foundation models are big, self supervised models. If you're a machine learning nerd like me or you have machine learning nerd friends anywhere in your Twitter network graph, you've probably seen demos that are really incredible about these models learning from scratch to generate text or answers to things or images that are quite amazing. And the question on the heels of all this amazing progress in these foundation models scaling up that we ask is how does this actually connect to providing production value for our customers? And the answer right now is it doesn't. In most places that we see all this ability to generate exciting text and image doesn't really translate to enterprise automation. And we'll get into this more but we see that the data and the data centric development as the bridge to connect that. And that's what we're, you know, the core what we're announcing today and we'll get into more in our discussion.

Awesome. I want to get into like the role snorkel and the new product you're announcing today is going to have in actually putting these foundational models to use. But I want to step back for a moment and ask you to spend a couple of minutes just motivating foundation models. I'm sure, you know, all of us have played around with ones like, you know, whether it's GPT3 on language generation or models like Dall E on image generation. I know the first time I used Dall E it felt like magic. I never was good at drawing, so it felt fun to actually be able to create interesting creative assets. And so I think we like, we taste the power of them. Yet at the same time there's a question of like, okay, how do we go from these really cool demos to like these things actually changing the ways we work and live and operate and so maybe spend a couple of minutes on that Alex and like what's hype? What's real?

So first of all, I want to plus one that excitement about these models, right. I also have spent some time playing around with these, you know, these multimodal models that can generate amazing images and the text based ones and how they work is not fundamentally surprising if you've been watching the space. You know, we've used, I guess what now need to be called medium language models, but the same fundamental architectures and types of models in snorkel flow for four years now. Things like Bert Distilbert, et cetera are some examples of text that many of us run into and we use in platform and support in platform. But the degree to which they've scaled up based on the increasing amounts of data compute engineering work and the results are really amazing and exciting for the field. It's a really exciting time. Even on the academic side, I could say with my academic hat, seeing this shift from a very kind of toy theoretical kind of view of machine learning in many places to kind of studying emergent properties of these gigantic foundation models that we barely yet understand. It's an exciting time to be in machine learning and there's, there's real, there's lots of hype, there's always a number of hype, but there's real, there's real potential and real, real progress there. That's exciting. But I think the other side of this is that these are not anywhere near getting into production at large enterprises in critical use cases. And that's something we see, something we hear from customers and you know, they're very pessimistic about their ability anytime the next couple years to say, deploy GPT3 in a real high value use case. They're still, if you talk to a top 10 US bank or a government agency or you know, they're still working on deploying, you know, basic machine learning models from five years ago through model risk management and at scale. So how do we solve this gap and why does it exist? You know, I'll highlight two kind of core challenges. One is around what I call adaptation, the other one's around deployment. The first one is just a natural thing that anyone in machine learning is aware of in concept, which is that if you have these big foundation models, they're great at generative and kind of exploratory human loop processes, they're great at generating text and images, et cetera. But when you actually want to adapt them to predict or automate something at high accuracy with guarantees of performance, like you do need to do to ship in the enterprise, they need adaptation or most commonly some kind of fine tuning or prompting. And doing this for complex real production use cases usually requires surprise, surprise, labeled training data and the constant iteration and maintenance of that, which obviously is somewhere where we focus and have focused for years with, with, with snorkel. The second thing is deployment, which is that again, getting these gigantic models into production is hitting and is going to hit in my estimation and the estimation of our customers, walls around cost, latency and governance and risk and bias for quite a bit of time. So the question we ask is how can we actually still use them and get all of this exciting value, but actually bridge into something that's a deployable artifact? And again, the answer that we've worked on for the last year and a half and are going to be announcing shortly is around again data centric AI and using them to power data development, but ultimately ship something deployable that can actually have an impact today in the enterprise.

So it sounds like customers are excited about the potential of impact of these models and they see the upside. Yet there are these challenges, as you point out, both on the adoption side, but and on the deployment side. What are we announcing today and what's the impact it's going to have on operationalizing foundation models in the enterprise?

So we're announcing a set of capabilities that are embedded in our existing platform, Snorkel Flow and the existing data centric workflow that we call the Snorkel Flow Foundation Model Management suite. And these are basically tools to use foundation models in a data centric workflow that actually can lead to production value today, even if you can't deploy a foundation model in production, getting the power of that and using it to accelerate and improve your processes. And it's exciting that we've already seen this in production with customers. What this consists of is a couple of features that we refer to as foundation model, warm start, prompt builder and finetuner. And the fine tuner one is maybe the simpler place to start because we've been doing this for years and now we're just offering support for fine tuning these these larger model classes. So getting a little into the weeds. One of the most reliable ways to actually get a model to high accuracy on a complex bespoke task for say an enterprise use case is to what's called fine tune it or kind of retrain parts of it parts of the model for a specific objective with labeled training data. And obviously, you know, our core value prop and our core, you know, technological contribution for years has been programmatically and automatically and iteratively developing that training data or enabling users to do that so that kind of fits immediately into our existing workflow and product. And then the other two features are all about how can you use foundation models to accelerate the auto labeling and development of training data, even if ultimately you're training a smaller model that'll actually be able to be deployed in your existing mlops infrastructure. And these features really came out of both research side where there are several papers on the academic side we posted from Snorkel from Stanford over the last last 12 plus months. They also came out of customer questions really customers asking us, hey, we have all these people who are excited they're playing around with GPT3 on their laptop. We're not going to ship this anytime. I had a customer we had dinner with at a top 3s bank we work with who likened his attempt to get GPT3 through model risk management as a Don Quixote like you know, tilting at windmills activity that he was doing just as quote unquote art. So how do you actually get this into production for use cases that matter to the enterprise? Well, part of our core view and what we're releasing with this suite is the ability to use foundation models to auto label data to both jumpstart the labeling. So you can start just in circleflow. You put in the class names of what you're trying to label and you get basically instantly get up to the baseline of what some foundation model can auto label. You know, some classes that may magically get right foundation models do that. For some of the lower hanging fruit. Others it's almost certainly not going to because these foundation models don't just work out of the box. It's always dangerous to draw analogies between AI and humans. But in this sense, think about a generalist who's read lots of Reddit and learned basic English and then asked them to, you know, parse an insurance document or a medical report or a complex financial document. Right. People need specialist training. So foundation model is not going to get you these trickier parts of the problem out of the box. But our warm start capability allows you to through zero shot, what are called zero shot techniques get there kind of right from the beginning. So it jumpstarts your progress in building your data set, building your models. And then our prompt builder allows you to prompt using natural language or code templates, these large language models to help auto label data in targeted ways. But then what you can do is you can also add in other approaches for labeling, you can use our iterative workflow to correct the errors and ultimately you can distill all this into a deployable model that might be 10,000 times smaller and also is going to be more accurate at the target task, which we're also releasing some, some case study results about as well. So this is a way to actually get this foundation model power bridged to complex enterprise use cases that could actually ship to production, which we don't see anywhere today in terms of existing solutions. So we're excited to fill that gap, which is quite big and existential for an urgent for enterprises who want to stay up to date with AI with these capabilities.

Just to make it concrete for folks like Alex, is there a good customer or use case you could talk about? I mean, as I hear you characterize those different aspects, the things going through my mind are how much faster I can get my data ready for machine learning, the impact and performance of the models I'm using and how that might drive business performance around my use case. Just overall accuracy and performance.

Let's dive into that. And there's more that if you're interested in the nerdy details, there's more in both academic publications and in some of the stuff we're releasing today in terms of case studies. But let me actually take a step back and just give one other anecdote that I like to motivate with. So I was trying to generate a snorkel logo. Our logo is a, is a little blue octopus wearing a snorkel underwater. So I don't think, I don't think there was enough support in maybe the data distribution for something as ridiculous as that. I couldn't actually get, I could get scuba gear on an octopus. I couldn't get a snorkel underwater on an octopus. But I got some pretty amazing, astonishing images and it took me about 30 tries. So I went through 30 samples, kind of tweaking the prompt and just generating samples until I found one that you know, looked reasonable and it was a pretty awesome result. So for an exploratory kind of generative human loop process, that's an amazing experience and outcome. It's awesome to see this boom in creativity around shaping those workflows and harnessing that power. But now if you shift the enterprise mindset and to the framing of most enterprise problems, which is automating or predicting or classifying something at high accuracy, that took me 30 tries. That's a 3.3% hit rate, right? That is below unacceptable to even consider shipping to production. So how do we bridge that gap? Well, let me give some, some examples that have actually been, you know, built using these, these, you know, the beta version of these capabilities. So one setting is at a top three US bank that's a customer of ours, some internal experiments around AML KYC problems. So this is pulling dozens or hundreds of pieces of information out of complex, highly variegated customer documents that then power the, the anti money laundering, pardon me, know your customer process. Going back to a customer example, one example now is, you know, one of our customers, a top three US bank on an AML KYC application. So I'll obviously obfuscate some of the details, but at a very high level. This problem which we previously supported on Snorkel flow without foundation models helping as directly consists of pulling out all kinds of, you know, tagging or extracting all kinds of pieces of information from complex, you know, multi hundred page customer documents. Very variegated, very complex and messy. It's not a standard form or a simple problem. And by applying these, these warm start and then prompt builder technologies to it or approaches to it. The really exciting thing first of all is you could use warm start and basically go from no labeled training data to label training data and a resulting model which is, you know, in this case the easy part that actually got above the performance bar, you know, above 90% accuracy or whatever the specific, you know, I'm obviously going to obfuscate the specific number that they hold as their, their internal performance bar, let's say 90, 90% for a handful of these, of these extractions, you know, what's the customer name, what's the, you know, what's the execution date of this document, et cetera, some of the kind of simpler types of things we were trying to label and then quite predictably some of the more bespoke and complex tasks about maybe 80 I think about 80% of the ones that were that they were trying to extract for this task. You know, the foundation models kind of basic approach, a zero shot approach got 30, 40, 50% of the examples correct. So nowhere near production quality, but a really powerful way to kind of jumpstart your efforts. Then what you could do is you can prompt these large language models so literally you're using natural language prompts to auto label along with other more kind of surgical approaches using regular expressions or patterns or other things we call labeling functions inside snorkel flow and have a iterative development environment for and within a few hours got all these up to above the performance bar. This is a process that previously would have taken eight to nine figures worth of time in legal fees to do by hand or even just to label the training data previously in snorkel flow would have taken days or weeks. Now it's down to hours using this kind of acceleration from these foundation model techniques. So that's one exciting example. There's others especially it helps when you have many, many classes you're trying to classify. If you look at kind of toy examples in the space, and I don't mean that aggressively, it's just most examples that you see out there of using foundation models for classification prediction are from a machine learning perspective, very toy. You know, take these restaurant reviews and classify them as, you know, happy or sad. You know, classify these customer questions into one of 10 categories right out in the real world where there's, there's real complexity, you often have settings where there might be hundreds or thousands of classes. Right. And so another area where we've seen it help is using this warm start and prompting to auto label huge numbers of classes and then be able to make targeted iterations on the classes where there are still gaps. And again, this is a massive accelerant for these very big, what are often called high cardinality machine learning problems that we see a lot of as well.

That's a great example, Alex. And as I hear you talk and correct me if I'm wrong, it feels like foundation models are really an accelerant of the data centric paradigm to machine learning application building that snorkel brings to customers. And they're an accelerant across vectors, right. On the data management and preparation side, they're a new approach to much more quickly get my data labeled and in a state where it's ready for machine learning. And then as I walk down the kind of lifecycle towards deployment, leveraging these things can boost accuracy and just drive better and Better results. But what's kind of interesting is I hear you talk, because it's like there's really exciting new things happening. And yet the paradigm shift that, you know, we've been pioneering at Snorkel now for many years early around shifting kind of from model centric to data centric, if anything is more pronounced. Like you have these much more powerful models, but still like to get them to the place where they're performing at the accuracy levels and with the controls and guarantees that you need in an enterprise context. It all comes back to the data.

Yeah. And even just taking a step back from specifically what we're announcing today, it's worth noting that these bigger ideas we've worked on, of course, I'm a little biased here, having worked on data centric AI and this push for the last seven or eight years, but really we view this whole area of foundation models as being intensely aligned with this idea of data centricity. If you look at the latest progress, I think there was, I was looking at the OpenAI whisper model recently, the paper there, the notes on stable diffusion and some of the, we have some collaborators with the Lion AI people who are working on some cool clip stuff. The architectures of these foundation models are not changing that rapidly. They're, you know, there's lots of innovation, of course, but they're fairly standardized. The engineering efforts are, I think, incredible. But a lot of the effort and a lot of the differentiation you're seeing is how you curate the data sources for training these models. So all the way from the start of building these, these foundation models to fine tuning them for, or adapting them for specific use cases, it really is all about the data. That's the really unique vector here. And there's a lot of work showing that recently, some of which is from us and from others. So first and foremost, this biggest concentric circle of what we care about at Snorkel, this idea that AI is data centric, it's about developing your data, it's very central to how foundation models are working and advancing today. And then down to this specific idea, the Snorkel flow platform of being able to programmatically label your data and iterate in this data centric way. You know, it fits perfectly there in our view because you can use it to leverage the best of these big gigantic generalist models while really sharpening and iterating and fine tuning them into, you know, the kind of specialists you need to actually perform at high accuracy in production, which is the final output we produce.

Awesome. Alex, We're Spending a lot of time talking about, like, the upside, potential impact of these models. Let's kind of talk a little bit about the other side of the ledger, right? Like, what are the risks around using these models? And when you talk to your customers, what are the things that. That's top of mind for them. Like, one of the things I'm interested in is like, do things like explainability and actually auditing model predictions change in this world and how do customers, like, capture the upside but also mitigate any risks?

Obviously a lot of our approach, which is very oriented towards using these foundation models to assist in the development of smaller models that may emphasize explainability and interpretability and guarantees more stridently, is based around the challenges that we're all facing here around these things with foundation models. Even the way that we've studied foundation models on the machine learning academic side has changed. I was up at University of Washington yesterday where I'm on faculty, and was talking with some folks there about how it's really changing machine learning in many ways from a, you know, very formal kind of field, very, very theory and formalism heavy, you know, guarantees about explainability or bias, etc. To a emergent property kind of empirical field where you just kind of poke at these things and see what comes out. A lot of the research about foundation models is like, hey, I, I tried this prompt and if I say, you know, think really carefully about your answer before the prompt and it works better and isn't that neat, right? So we're really just starting the process of kind of poking with sticks at these gigantic things and they're really fantastic emergent properties. Now from a governance and risk and bias perspective, you know, that is and should be a little bit terrifying. And that's why, you know, we see the blocker. It's not pessimism about these, these models, it's just the reality that it's going to take a while before they get to being served in production before someone, you know, I mean, this is a little bit of a glib phrasing, but it's going to take a while before let's say a, you know, a large us, you know, company plugs some model that was trained on, you know, Reddit directly into, you know, their, their customer facing Chatbot, or I guess that has happened in certain experimental tech company phrase, to predictably terrifying results. So we have a long road to figure out how to really explain, interpret Debias and put governance controls and guarantees around these foundation models. And that again, feeds into our fairly unique way of using them, which is harnessing them for a data development and data centric process that can be separately audited and controlled and explained and de biased rather than just plugging them right into production. So really our excitement here at a company and a technology level is, is filling this really strident gap between this amazing surge of foundation model progress that we see on our Twitter feeds out in the public domain and real complex high value enterprise use cases. We see snorkel flow, our platform as that bridge, we see data centric AI development as that bridge and we're seeing it already working with customers as we went through some stories of however, I'm curious to hear you Sam, what do you see in the broader landscape from your pov? Because it's pretty exciting out there right now.

Yeah Alex, it's not going to surprise you that if I look at both founders who are coming into our offices and sharing more about their companies and then customers later today I'm hosting a large bank and I looked at the topic list and number one on the topic list they want to talk about is foundation models. I think foundation models have really taken the zeitgeist, not just in AI, but in software generally. And certainly our view is like we are early days of a new computing trend and literally everything we do ranging from how we live but importantly to how we work, business workflows, these things are all going to be transformed by the advances in AI and around foundation models in particular. We've crossed the chasm now on tractability where now we're seeing these results both on the generative side and, and on the discriminative side where it feels real and like people can't yet fully connect the dots and be like okay, I get exactly where the business value is. But they're like no, no, no. Clearly this stuff's performance has like tipped. And importantly it's just getting so much better as we keep throwing data at these models that like this is going to deliver on the promise of AI. And like in many ways we've been investing in AI for a very long time at Greylock, Alex, as you know and I think if you go talk to customers, for example, most CIOs have been thinking about AI for many, many years now. But I would say and I'd be curious your thoughts. Like most people probably feel like we're still in the trough of disillusionment, right where they've been told this promise of AI, but when they actually look in their environments and say where is AI having an impact on my business that story is still thin and I think there's a number of things that will be required to tip that story. But that story is going to tip really fast now. And I think foundation models end up being an important part of that. So we're like very, very excited now. I think we also are sober around. It's one thing to show a really cool demo and it's another thing to use your parlance, adapt, deploy and utilize a model in production that actually drives business value. And I think our view is how do you take the advantages these foundation models have, right? The scale of pre training that goes into them, the characteristics around transferability of performance. We don't believe for most use cases where performance really matters, you're going to have this kind of one magic model to solve everyone's pro, you know, solve it all. Like instead what's going to happen is you're going to take these models that and take advantage of the scale of pre training that's happened and then really train on top and fine tune to your use case. And so if I'm a large bank, I might have my own foundation model and maybe that's bank wide, maybe that's use case wide, right? Like maybe I'm looking at a specific set of kind of financial documents and I want a foundation model just for that in the context of my environment. And I think that's when those models get trained and deployed, that's when we really see the power of all these research advancements actually have kind of enterprise impact. And you know, candidly that's why we think actually snorkel and platforms like it. But you know, can be the missing link or bridge between the potential these foundation models have and actually realizing their impact in production. And like that's why personally I'm just so excited for our new product to get out there because I think we could see dramatic acceleration of AI impact across many, many enterprises.

Sam, it's really exciting and I think it's a, I won't even pretend that it's a surprise that there's concordance in our views because obviously. But you know, I do want to kind of just, just touch on a couple things you said just to really plus one them but emphasize what we're seeing. You know, number one, this tremendous uptick in excitement. You know, we're seeing that it's, it's driving incredible and exciting momentum in, in our inbound, in our business. And a lot of it is around these, these features that we're now publicly announcing but have been previewing with Prospects and customers for in some form or other almost, you know, six, six, nine months plus now. So it's, it's very exciting, I think on both sides for customers and potential users in the enterprise of AI, but also for, you know, for vendors who have something to say about foundation models like we do. Number two, there's also definitely a healthy spoonful of this, this skepticism and disillusionment. Right. We've seen this before and this is where we've lived and still live around the blocker of training data. Right. Enterprise spends a huge amount of money building a center of excellence, a, you know, well heeled and well skilled team. They've got all the fanciest machine learning state of the art models that are out there often in the open source and then they're blocked on asking some line of business team to label data for 612 months before they can train a single model. We've seen that disillusionment and we've helped to solve it. That's been our mission so far. We're seeing a similar kind around foundation models. Tremendous energy, excitement, but can't get it to production. How do we do that? And again, we think data and snorkel flow can be the bridge there and we're already seeing it happen, so it's exciting. And then the final thing to add on is that I completely agree with. Well, it's kind of baked into the name and the nomenclature that both of us chose for this podcast. Foundation models, right? The idea is they're foundations to build on. You still got to build the house. And there's not just one prefab house that's going to work or building for every single use case. That's that, you know, that's where all of the, the adaptation, the kind of specificity and all the, the really detailed design specs have to happen. So if you think of foundation models as, okay, they're going to be this magic out of the box solution. You know, you're living in a fantasy world. You're living in the, the, you know, cherry pick demos and toy problems world. But if you think of them as just incredibly strong and powerful foundation that you can then do all your building on top of in a massively accelerated and really, you know, promising way, then you know, I think you're on the right track and we're here to support that with snorkel flow.

I love the house analogy because I think that absolutely nails like both the potential, but also what has to happen to really get these things operationalized. Alex, we've Talked about like today's really exciting product announcement, but it'd be remiss if I didn't ask for more like what's next and what are the next couple things coming on our roadmap?

Well, unsurprisingly, a lot of them actually have to do with foundation models. Right. That whole, you know, foundation metaphor really extends. But a couple things that are worth noting. First, we're doing a lot of exciting stuff around image and video that's having some exciting customer impact. You know, we have a large E commerce company we've been working with who with our beta product for image was able to, I think take a four to six plus week process and now do it in a few hours at greater quality compared to, you know, manual labeling. It's all about programmatic labeling, data centric iteration. It's also powered by some of these prompt builder and warm start techniques that we talked about. So going into some of these exciting new modalities where we've done research for years but haven't GA'd yet in product that's coming out soon, we're really excited about it. And it does also connect to foundation models, another theme which is a, you know, a little bit different, but it's super important and it ties in, of course to everything we talked about here is ways of empowering and connecting to what we call subject matter experts and annotators. The people who actually know how to label the data, know how to validate it, know the business value that is attached to the model's output. Right. These folks are critical. If we take even a step back and say what's, what's one indicator that an enterprise is not going to successfully adopt AI, I'd say it's over siloing and separation of the data. Scientists and the subject matter experts, whoever they may be, lawyers, underwriters, line of business analysts, whatever the specific title, these two types of canonical roles have to work in lockstep to actually get AI developed and deployed to production. And we're really doubling down on our suite of collaboration tools to support that. That's whether you're doing this to validate or correct the foundation model with your subject matter experts help or just get some manual labels for validation that collaboration is critical and it's a big investment that we've been making and are continuing to make with partners. So maybe those are two themes I'll.

Highlight exciting themes and a lot of good stuff coming on the roadmap. Alex, this has been an awesome conversation and I come out of it just really excited about Foundation Models, Snorkel's role in foundation models and the impact we're going to see them have across across enterprises and businesses. Maybe as a closing question we talked about roadmap like what can people get their hands on today and how do people go start deploying Snorkel to leverage foundation models in their environments?

Yeah, well we are really excited today to be announcing again this Foundation Model Management suite. Three key features there the Foundation Model Warm start that gets your data auto labeled and ready just by putting in the class names using the magical power of foundation models. Foundation Model Prompt Builder that helps you use foundation models to auto label and iteratively develop along with the full suite of other approaches supported in Circle Flow. You can use them all together and then also you know the foundation model fine tuning where if you can use foundation models in production you can use that. Otherwise you can use the existing suite of models that are often much smaller and more deployable to kind of get that out there. So those together are just extra features in the existing Snorkel Flow platform workflow. We're going to be sharing a lot of information and some demos today. There's going to be some upcoming demo walkthrough sessions where we go through the workflows that you can now be empowered to do in platform coming up in some sessions and then if you're interested in getting hands on demo the platform just shoot us a note request a demo. And we're really excited to get this out there and actually be this bridge between this amazing progress in the field of foundation models and real complex enterprise production use cases solving the adaptation and deployment problems. So yeah, come check it out on our website.

Awesome. Alex, thanks for joining us today and excited to get this out to our listeners.

Thank you so much for having me.

That concludes this episode of Gray Matter. If you like what you hear, please subscribe wherever you get your podcasts and please be sure to leave a five star review. You can also find us on Twitter reylockvc. I'm Heather Mack, thanks for listening.