⚡️Every product of the future will be a living system — Ronak Malde, Trajectory.ai

Okay. Uh we are in the remote studio with Ronuk from Trajectory. Congrats on your launch and welcome to Len Space. >> Thanks so much. Yeah, great to be chatting with you. Thanks for having me on. >> Yeah. Um I don't think I I like we like quite over overlapped when when you were at Windsurf but obviously I was close to the windsurf uh team Verun and the other guys and it's really surprising to see that uh you know you have sort of gone through that journey at deep mind and come out uh with such a strong thesis on what's next and continue learning. >> Yeah. No, it's it was a great time honestly at at Windsurf and it's crazy to think that even two years ago the world just looks so different. I I still remember when we first launched Windsurf, it was like Sonnet 3.5 had just come out and we're playing around with the capabilities and everything. Um, and I distinctly remember there's one time where we had started the IDE, we had built the agent out and everything and then we all started like circling around Verun and he was like he told the agent to like make 2048 and it was just like silly game whatever. We like we saw the agent and it just like first shot made the game and that was like that mind-blowing experience. I mean obviously now everyone's experienced it but getting to to build the first systems like see that in the room it was just so magical at the time and then obviously getting to ride that wave to to now deep mine and everything it was super cool. >> I think Kodium and then Windsurf had a bit of a mix of like product engineering and model training. what was your background like coming into this? >> Yeah, so I had just finished up my masters at Stanford. Been doing research in in AI for for quite a bit of time. That's actually where my I met my co-founders as well. Uh we we met like literally first week of freshman year at Stanford, which is super awesome. We can get to that story after. So then I I joined Kodium. I actually had the option to go to Google for to work on Gemini post training. That was kind of the status quo. But I met the team there and I was I I realized just it's it's all about the team. It's all about moving fast and getting to see that at Windsurf. It was incredible. So took the vet on that team, joined one back when it was called Kodium. And I had this thesis that it was either going to be math or coding, some sort of structured information that gets us to the next leap in AI. And it turns out that coding is a lot more economically viable as well. And so that's basically how I joined the team. I started on on the research side there. So initially we were training autocomplete models at windsurf like back in the kodium extension. Afterwards we started to realize okay we're getting so much user signal not just like accept reject or what people are typing in the code but actually from the agent right and like how are you building these entire websites entire apps and and so we sought out to build our own foundation model to capture all of that kind of user signal and and really get the flywheel unlock for windsurf. And so I started working on this model called sui 1. Uh you might have seen now there's like later iterations like sui 1.5 and 2 cognition is working on and this was the kind of major unlock for the company as well is we had all this massive data. Um we were able to post train on all of that user signal uh and now beat the frontier. Um so that all happened and it was it was super fast cuz then right after that all the Google and cognition acquisition happened I think like 35 days after or something. So that was that was super fun. >> Yeah, we've done episodes around that just because it's a matter of just general interest. Uh what what is your take on that day that it happens? >> Yeah, it was it was crazy. I actually don't know if anyone's told this story, but go ahead. So I I guess yeah I was on the research side and then if you remember at the time like we everyone thought that OpenAI would acquire Wind Surf like the the news had come out and everything and and so like we we get a a text before from our CTO Douglas and he's just like meet me in this hotel conference room like tomorrow morning and don't tell anyone else. I'm like this is so suspicious but we go over there and like I fully expected it to be uh like Sam Alman or something right and it was going to be okay. You thought you were going to like it was like it's closed like party time, you know? >> Yeah. Yeah. I mean, I thought like the news like everything had been leaked that it was going to be open AI acquiring us. And so I Yeah. We go there, we thought it would be open AI and then we go there and it's it's like Demis on the call and then like all of the like deep mind Google people. And so we're like, "Okay, I guess we're going to Google then." And it all it all just happened super fast. Like that was on a Thursday morning. Uh that same day we just we gave up like all of the like badges, computers to win surf and then the following day we were at Google like full employees or like at Deep Mind >> and so it just it was just like insanely fast and and like the news came out I think Friday morning. So uh yeah, it all just happened that week. >> Yeah. And then Cognition bought the rest on Monday. >> Yeah, exactly. >> Yeah. [laughter] Uh yeah, very very crazy. And uh it's very cool to hear that Demis was involved in the deal. I I know that like I think there was like several people like senior people like Sergey um pushing for it and obviously like it it's kind of like a aqua hire but also I think they had to just pitch the team. There was like 30 people or something on like okay what are you doing at Deep Mind? >> Yeah it was um there were a lot of execs involved. So Sergey as well he's he's like the main guy that was super excited about Windsor. Um, he was actually on I don't know if you saw the anti-gravity launch video that came out, but uh he was a a main character in that as well. And and I think for Google and DeepMind, they're also starting to realize that you need to not just own really stellar models, but you also need to know what users are actually doing with products and like seeing all of that signal, all the stuff that we were capturing at Windsurf and being able to like really close that loop. And so then, you know, we we went over to Deep Mind and launched anti-gravity got to contribute to Gemini 3 as well. And it was really awesome like to see what a powerhouse of a company with a ton of compute can actually do and and really really leverage just like cutting edge technology. And I think now everyone's starting to realize that you can't just train on like offline benchmarks or like scaling up human labelled data. You really need to understand like what people are doing in the real world especially as AI is becoming more and more prominent for real world use cases. >> Yeah, totally. I mean and that thesis clearly is panning out today even with Cursor and XAI doing a similar deal. >> Exactly. Yeah. >> So, yeah, I'm happy to fast forward to when was the moment you decided, you know, you have a Kushy job at Deep Mind. I'm sure they are paying you like millions of dollars to just do what you're doing. Um, where's the moment you're like, "Okay, no, like you know, it's uh it's time to go start this thing." >> I Yeah, obviously the acquisition for was for $2 billion and went over to Deepine and then I decided to give up all the acquisition money to start trajectory and it all came down to so I obviously we've been having all of these innovations and coding and and seeing that real world usage, but it's very clear now that we're getting that explosion for every single other domain, right? like legal, healthcare, finance, they're all going to be supercharged in the way that coding was for the last few years. And the other part though is as we're getting this kind of transformation, it's still that AI like it's it's obviously very powerful, but it kind of acts like any other software uh in that it's very static. Uh like the model that you used yesterday, it's going to be the same model making the same mistakes tomorrow. All of those corrections you gave it, the edits like in any product is all just being put to waste. And and I think I realized that okay there are a few technologies like we did at su 1 like cursor composer even cloud code uh that are starting to build the model around the product and what people are actually doing in the real world and this is that compounding loop that allows them to pull ahead that is what I want to bring to every single company that power of continual learning as people now refer to it and and so you know I I got together with my co-founders I mentioned they're from Stanford we we at in the math class in in first week of freshman year and uh they went on to do awesome stuff in the research world as well. So Michael was at deep mind working on he was in deep mind proper for a long time working on robotics there. So he was on the Gemini 1.5 robotics release and uh leading a thinking team there and then Arjun was on the vision pro for a long time. He he was working on the core interaction models and training those shipped those day one to to all the vision pro users. Um and so in some way or another all of us were kind of working on how AI interacts with the real world right through either coding robotics with ARVR. And we realized continual learning is kind of the ultimate like paradigm to do that is like how do you have humans in the loop? How do you build this intelligence around them that is constantly learning and growing on its own? And I think that's going to be the next major unlock of AI is is capturing all this real world usage and and really going with it. Yeah, I I think that's very exciting to see. I think broadly very true. It's weird because all three of you went to such like notable companies. It's almost like you planned this like you're like, "Okay, we'll split up for a while and then we'll come back together." Yeah. So, so okay, you you decide to work on this problem and I guess what's the approach like you know like I I think continue learning is a hot topic among a lot of folks, but I think people approach it from different angles from like the train it into the weights versus just have a bunch of skills and markdown files or whatever. you really see Laura recently just I guess like just just tell people more about like maybe with the schools of thought in continued learning and and the one that you're choosing. >> Totally. So I think it's helpful to actually like we've been talking about just like the research and and stuff that like enables this. I think it's useful to think about this from the customer's perspective. So take a company like Harvey for example. They they have a ton of complex legal workflows where agents are trying to research documents and then redline a contract. And unlike coding where we can obviously a technical audience can just like guide the agent in the right direction, we're able to be tolerant to mistakes for a field like legal like getting 80% of the way there is the same thing as zero. And and this is for a lot of fields where they have very expert data of what people are doing. Maybe an agent attempts something and then the user needs to go and correct it and get the rest of the 20% of the way there. That's that useful signal that we're able to learn off of. So actually we uh we I I can share my screen and show some of the work we've been doing with Harvey. But uh I guess before I get there, the the high level of what we've built for all of these companies uh is a platform for continual learning. We we essentially take all of the data, all of the expert traces, the agents and distill it into one format which is what we call the trajectory. And it's basically all of the data that is needed to then create the evals, the judges, the environments, everything, all the components that you would need for training. And then we're able to create that end to-end loop of a self-s serve platform to optimize the agent, the model, the harness, everything. Right now, we're starting with the model. And that's what the the major kind of alpha is for every company is being able to own models that are faster, cheaper, better at the frontier. And so I can share some of what we've done with Harvey, which is really exciting. One of the the main things that Harvey really cares about is obviously they're a regulated industry. They want to have sovereign intelligence as well. Um and so we partnered with Harvey and Nvidia in order to train Neatron 3 super uh in order to get to the paro frontier. a lot of the legal workflows that they care about. I can actually fast forward over here where we were actually able to capture with along with Harvey's lab uh data set a lot of their expert legal workflows. And so um the things that they care about for example are issue spotting, analysis, explanation, citation, reference, completeness coverage um etc. and all of these things the they have sort of like parallegals and lawyers uh going in-house and working on mission critical tasks and so with our model we were actually able to improve these sort of metrics across the board and then also the the really cool part is Neimatron is a drastically cheaper and faster model than the frontier and this matters at scale when you're deploying to not just a few users hundreds and thousand hundreds of thousands of users across the board. And so this is something that's very exciting for the legal domain and something that we're able to do on our platform. The other exciting part is because we're building this I think essentially like every company has their data sources in different formats. We're able to build this as a scalable platform that gets all of that data without any kind of forward deployed work and really improve the model. And and the cool kind of proof point is we started out with working with a couple companies. Our our current partners are Clay, Harvey, Rogo, Dakugon, Mor. The first engagement, it took like three months to like set up the entire thing. We're kind of building the airplane as it was flying and now we're able to train a model with Harvey in under a month. and then onboarded a new customer and within a week we were able to train a model and really get the flywheel going. Um, so it's it's really cool to see just like how powerful continual learning is in practice. We're obviously just getting started and I want to be powering kind of every company that that is pushing the frontier and knowledge work. >> Yeah. Yeah. Congrats on this. I mean a couple observations. First, I think you're the first team doing this I've encountered using Neimotron 3 Super. Any comments on them as a base model versus you usually, you know, people who choose like one of the Chinese models? >> Yeah. Yeah. I I mean, so I think it's really awesome to see the open source community just in general. Like obviously China has been like killing it with models. I remember back when we were at Windsurf >> Yeah. 31 is now known to be a post train of >> well and I think back then like before in like 2024 or so like open source was so far behind and you had like some Quen models you had Deepseek 2 and it just like didn't seem promising and then all of a sudden Deepseek V3 comes out and everyone is like wow the they're using Frontier Paradigms they obviously open source or like showed the world GRPO and I think since then the gap has and closing and closing over time. I think America or the the western world has some work to do still like obviously having one trillion parameter models like Kimmy like and and amazing models like GLM and Deepseek. Uh I don't think we're quite there yet for that size of model. Uh but you can already see like um I mean GP2s was kind of one of the best base models a year ago that OpenAI put out in the Western world. Like since then, Neatron has gotten significantly better. And keep in mind, this is a 12V model, so same size as GP2s. And and so I think as time goes on, like Nvidia is investing a ton of money. The the world is uh is gearing up for I think a lot more open source models in the Western world as well. Um which does matter like for legal, for finance, for healthcare. And and so I think this will be a leading paradigm. And so we wanted to be ahead of the curve and and really make that a powerful part of our platform. >> Got it. I'm sure Nvidia will be happy to to hear that. >> Uh the other >> Yeah. I I think the other thing also is I so for me the general question of like why people haven't pursued you know let's call it online models or continue learning is like just the curation of the data like what what is the sort of data curation solution here? >> Yeah. Yeah, totally. So I think the the interesting thing here is like every company is storing their usage their their data in slightly a different way, right? And you have this entire system of basically uh two things. One, what are agents doing in a product? Um so you can imagine for cursor they're obviously like writing code, they're running commands. Uh and then for like I don't know a GTM workflow, it could be that the agents creating tables or enriching columns and then so that's the easy part. That's like in a open AAI like tool call tool response format. The interesting part though is what are all of the other things that users are doing in a product. And so at Windsor for example this was users are modifying the code after an agent. They're saying like hey I actually wanted this button on the left after the agent wrote a website or like I wanted this class structure this way. That's all the very useful information that's happening in a product. I I think when people think about like online learning, continual learning, they'll first think of like like accept reject or thumbs up, thumbs down, like some of those like kind of binary signals. Uh it turns out that that's actually like very noisy. You can imagine like when is the last time you put a thumbs up thumbs down in cursor or cloud code never, right? And everyone's kind of vibe accepting everything. So it turns out those signals are actually quite noisy. But it what is useful is all the directions of like the user is saying hey you should have gone this way or like modifying some of the attempts of the agent. And the cool part about this is this works in any product. So uh for a legal workflow it's the same thing. the agent attempts to do some research and then write some red lines or create an Excel spreadsheet and then afterwards the lawyer goes back and maybe has to do some more research because the agent missed something and that sort of taste and then again also like modifying some of the contracts and so now the tricky part is how do you take all this data and then condense it into the right format to train on and so there's a couple of really exciting things here that we've been working on. One is just purely an infrastructure question. When we figured out a way to take all this information and build an auto research pipeline that is now able to take like whatever these vectors are of what the user is doing and make those into judges into Eval's environments. Um but the other exciting part is the actual algorithms behind continual learning. Um, so one of the other really cool things I I'll just go ahead and make share my screen again uh that we released very recently is scaling updo self distillation policy optimization. So if you're not familiar with this I I can give the high level real quick. So for RL basically the way RL works and that's kind of the leading post training paradigm is you have an agent and it goes and explores the world a bunch of times and then at the end of the day you get a reward signal. It could be a judge, it could be like something from production and then you basically take that number and then use that to update and say like hey this entire tra trajectory was really good. this entire trajectory is really bad and then it learns from that. So it works pretty well. Obviously all of our frontier models that we see today are trained on RL but it doesn't really work for this new world of uh continual learning. The reason is RL it's still taking all of this kind of useful information from the real world like I mentioned all the corrections and everything and putting it into just one number which is really broken. And so what self distillation does which is really cool is takes or I guess the high level here is we start out with a student roll out. So if you're familiar with distillation in general it's you have a student that is usually a less smart model and then a teacher which is a smarter model and you're basically saying okay let me distill all the smart stuff from the teacher into the student. The cool unlock with self-distillation is you don't actually need a smarter teacher. Your student could be the smartest model, but to make the teacher smarter, I'm actually going to take some privileged information or a hint uh and put that into context of the teacher and now I have a suddenly slightly smaller smarter model that I can fit into. So in practice, what this looks like, let's take this example here where we have an agent, it's asking or a user ask like how much is my flight ticket to New York? we have the agent like look up some stuff uh and then we we actually get like the wrong ticket information. Now we can actually go back because from production we have some hidden information and we can say actually uh let me give the teacher a smart hint. Then we match the student log props to that uh teacher information and then suddenly we're able to take not just like a binary reward but truly like actual text and guide the model in that direction. So this is a huge unlock of STPO >> and we've done some very exciting kind of modifications and scaling it up. It's been done in a lot of academic cases but no one's actually been able to scale it up to real world use cases. So we started training on Apex agents which is a an awesome benchmark that Maror put out which is modeling a lot of real world behavior and we're able to see very amazing gains on some of these workflows. So for example also the convergence rate just like goes up way faster. This is normal uh gpo style training and then we're able to obviously be way faster. So this is this on our blog post as well. It's very awesome to see and we do some modifications as well to training on real off policy data which is like what happens in production not necessarily from your current model and making it a lot more robust to real world use cases. >> Yeah, that's this is incredible work. I can't believe you basically shipped like research alongside of your products and it's a it's a team of three. >> Well, so I actually we we have 11 of us now. So we've been growing really fast. We uh >> Yeah. You have a nice office. >> Yeah. Yeah. Yeah. Um I I think in order to build something really awesome in this space, we we think of ourselves as a product company and a research lab. Uh because I think it requires cutting edge research. The current algorithms aren't quite there yet to get to true continual learning. And so that's a lot of what we're exploring. But then it also requires deeply understanding customers and like building a good interface. No one's really made the product for when intelligence is just getting smarter every single day. And I think there's a lot of very interesting interface questions here. So, but the team is really awesome. We we've hired an amazing research team from OpenAI, from Meta, Super Intelligence, from Amazon AGI, and then obviously co-founders from from Deep Mind and Apple. And then also really growing out the product team as well. So, we we have someone on the team from Stripe who is building a lot of backend SDKs and then uh someone from Figma as well. uh really thinking about these interfaces and designing them. So I I think it's a concerted effort across both product and research to get here. >> Yeah. Let's go through more of the core IP because I don't want to cut you short. You you were just getting into some good stuff with SDPO. You have like this like 5 days of trajectory which is also like very very strong launch. I think one of the strongest I've ever seen, you know, in in this space. Uh yeah, keep going. Like don't let me [laughter] >> Yeah. Yeah. Um I guess I can share one more as well. And uh yeah, it was super fun. like we've just been doing so much across research and product and so it's really awesome to to be able to showcase that to the world in in our 5 days of trajectory. So I I can share one more as well is uh I mean our our goal is trajectory is to really empower every single company to just like get to the world of continual learning. It's such an exciting unlock when you can just put a product out there to a couple hundred users and now the intelligence just grows over time. And so one of the other exciting things we did is is open sourcing a training stack for continual learning. And so this is uh in conjunction with Sky RL uh Berkeley's Sky RL lab um in any scale as well. And so so basically the the high level here is up until now with training it's very much like a linear like if you're ever kicked off a run with like slurm or or one of these like normal training clusters it's like you kick off a job it like starts to spin up a bunch of resources spin up the GPUs in order to do the sampler do the training and then you run the entire life cycle of the training job and then spin it down. This is like normally how training works. The problem is with continual learning, you don't really have like training explicitly starting and stopping. You're probably running concurrent jobs at the same time. And so the normal training paradigm start to break apart. This is also very similar to the thesis like uh like thinking machines like a few of these other like distributed Laura uh sort of companies infrastructure has had. Our contribution is to make this open source uh for every company. One of the things here is like okay so this is like normal life cycle of a job is like set up you do a bunch of sampling training and then clean up. What continuous Laura does is say hey instead of a bunch of linear jobs like this I can actually stack things together have one dedicated pool for training and one dedicated pool for sampling and then now I can start to put all of these pieces together and run jobs in parallel. And so the results are really cool. We ran this on several different scale up of experiments. And you can see that uh across the board, even as you scale up experiments, so like two concurrent jobs, we're able to cut the wall clock time in half. As this goes up to four jobs, eight jobs, we're able to run concurrency really well. And >> how come how come eight goes up? >> I can also share it here. Let me let's walk through here. So >> okay. Oh >> yeah. Okay. So the way this works is okay so up here is like serial training run right like you're doing a bunch of sampling rollouts whatever but if you have a fixed set of GPUs then you have to run the job sequentially after another with continual Laura so it is actually slower per training job uh like if you just care about like I just want run one to finish as fast as possible then you should be doing like completely serially because all the resets are going towards that but with continual learning what you care about is just like let me have all of these runs like learning on the job. It also might be that like data is coming in not quite sequentially like this but maybe in batches right you get new data from production now you need to have like run four suddenly take up more resources. So it turns into a lot more of a dynamic workload and you can start to see here like like eight concurrent runs for example it takes about the same time as like sequentially like three jobs would take. And so this also scales out like we didn't fully push the bounds of this in this blog post. Uh we're we're currently scaling this up to like 16 concurrent runs and scaling up to this was also on like smaller models as well like scaling this up to like 235bs. And uh and the really cool part is obviously like this doesn't degrade performance whatsoever. It's still the same algorithms. We're able to see the same gains with any any sort of normal model. And uh and we wanted to also make this very plug-and-play. So we we have some kind of instructions here of just how to get started using it in Skyall which is really exciting. So >> yeah it's really funny as a operating systems fan like that you see like very similar like scheduler type problems starvation preemptive scheduling all these kinds of terms like you just take a normal OS class you like then you start applying it to every every problem domain like this. >> Yeah. Yeah. Exactly. I I think it's almost like training up until now has been a very research problem and I think people have attacked it from a researchy perspective where you don't care about all of these like kind of distributed systems innovations. Uh and suddenly now like training is becoming more production right uh like the end goal of trajectory is that every company can be kicking off their own training runs like continuously learning updating and that also requires a training stack that is built with a systems level or engineering kind of mindset and so so this is just kind of the beginning but I think there's so many different insights from distributed computing that can be applied to here >> got it by the time this comes Oh, you will have uh sort of finished your your 5 days. Any other sort of overall tech bets or directions that you want to hint the audience towards? >> Yeah, totally. So, this where we're at with the product right now is we're able to take all the kind of noisy information from a company like all the valuable signal of what people are doing in a product, be able to train the model and then re-upload that and then have that like observability, the evals, all of those things. But right now it's still kind of like these single components of the stack and a lot of these things like we are we are kind of like managing or or looking at like kicking off the training runs. The next step after this is one I like it kind of what we think of as act two of the company is I want to make this legible to the customer. So just like the right observability tools, the right abstraction levels for evals and then second give them control. So like I I want to have the customer say like a PM should that's like managing an agent product should be able to say hey okay here's where my agent is really good here are the areas where it's still failing and now all I have to do is just like modify the model modify like all of the pieces and suddenly tomorrow I wake up and I have a smarter model that I can see in production all of those things like should be in control of the customer. So that's the next phase. Uh and then after that the most exciting part is right now we're working with these AI native companies I mentioned Clay Harvey Deca Gun Rogo um soon we'll be working with the kind of tech incumbents you can imagine like the air table notion those sort of companies the stage after that is fortune 500 and in order to get there we need to get to true continual learning that I can just have just the observability layer right like I'm observing let's say in Walmart like what are users doing what are all of the manual kind of process es and then start to say, hey, okay, maybe a model isn't even possible today if you were to just use an off-the-shelf sort of model, but I can now observe exactly what users are doing. I can start to see the patterns, build out an agent to harness a model that all works perfectly for that workflow. And now you can imagine any sort of company, their front using product should just dynamically be learning from their users and constantly updating. in order to get here as well tactically we're starting with the models but very soon I I want to be also improving the harness improving skills doing maybe even the memory layer as well I think all of those in conjunction are the continual learning solution yeah I know you mentioned that at the beginning as well and I I think all of those are super important and also the interplay between the two like no one really knows what it means to train the har like optimize the harness and the model together right and I think there's very interesting research there and what that looks like so these are all very exciting areas that We're we're like super stoked to get into. >> Is it worth focusing on coding alone as a verifiable domain or are you un unopinionated because you got legal or you got the other stuff? >> Yeah. So for us, I mean I I think that just like we saw in coding for the last 2 years, I think that's going to happen for every single company and and every single domain. So our thesis was basically okay like coding is moving super fast. let's accelerate the rest of these industries to first just get to where coding is, but then after that like truly transform all of them with continual learning. So that's why we're starting with also companies in different verticals to also build our platform so that it's agnostic to whatever domain you're in. >> Okay, cool. Like very ambitious. I I I think I finally get it more after talking to you and hopefully people listening along can can get a sense of the scope of the vision as well. Where should you should people point to for uh trying stuff out? Um you know, what are you looking for in terms of hiring that anything like that? >> Yeah. Yeah, for sure. No, so we're going to be growing really fast and it it's exciting to see that. I would say like honestly the the main place where all of the the research world lives on is Twitter and so following along on there and uh no and and I like I want people to also like engage and be excited about what we're building. I I felt like when we were at Windsurf again like doing uh just like seeing how developers were using coding agents and like being able to be really in tune with that uh it was super important and and for us that's the same way like I want to just be in tune with like how customers are using our product and how how continual learning research is is kind of scaling up and us being able to um at least contribute some innovations to that field and and we're going to be I think like the two areas for us are obviously see like research and and growing that team and product as well. We we want to be really robustifying our platform, scaling it up to not just five customers but 15 20 50 very soon. So, uh yeah, that that's kind of the next frontiers. >> You got five very good customers in there that are extremely good logos for a lot of people. Uh yeah, I'm excited to hear more and and see more, you know, and people can check you out for sure. I think I've also invited you to my conference uh AIE where we are doing a continual learning track. I was very cautious about doing it. I was like, I don't know if the field is ready. But with you guys and some of the other folks also working on continue learning, I think it's time. Even though if we don't exactly know what the end products will look like, I think there's enough test cases where like Yeah. like people are trying to put this in production already. >> Yeah. I I think it now is the time of continual learning and so it's super exciting. I I'm very excited to be at the the fair. >> Yeah. Yeah. All right. Congrats. Amazing launch. seriously like so well executed. I'm I'm actually like very very impressed as like someone on the marketing side, right? Like I'm like man that means a lot coming from you. Yeah. [laughter] >> I'm like okay like this is like the bar now for like how to launch a products lab, you know? So yeah, congrats and excited to see more. >> Thank you so much. Yeah, thanks for taking the time to chat. This is super fun.