Latest / Talk Commerce / AI Revolution in E-commerce: Big Sur AI's Vision for Democratizing Advanced Technology with Vinod Ramachandran at ShopTalk
Transcript
- Speaker 1 (00:01.934) How's your top commerce going? Good. Speaker 1 (00:10.838) Hmm Speaker 1 (00:19.722) Sweet. Speaker 1 (00:24.845) Sir. Yeah, absolutely. So we started Big Sur about two years ago. The background for this is anytime there is like a technology revolution, such as the one that we are seeing with AI, this is even bigger than anything that we have seen previously, it's usually the largest player in the ecosystem who gets to benefit from it the most because they have the resources to adapt and quickly move and gain the benefits of the technology. So if nothing is done, it simply widens the gap between, let's say, the largest players in the retail ecosystem versus rest. So what we set out to build is hey how could we build a platform for e-commerce that regardless of the size, brands and retailers could easily access the benefits of AI. So we are essentially trying to democratize access to AI and level the playing field so that every brand and retailer can compete with each other. Compete against the larger ones, right? And so the platform that we building is, we are effectively building a set of specialized AI agents for every function within a business. today, the question is not about adopting AI for just one part of your business. AI is fundamentally changing how organizations operate. And we know, for example, with all the economic uncertainty, teams are shrinking. Speaker 1 (01:44.716) CAC has been growing and whatnot, there is a huge opportunity for AI to influence every part of your business. So we are building that platform where every team in your organization should be able to use AI that can drive significant increase in sales, reduction in costs, and most importantly, amplify the team's output. And so we currently have three of these AI agents that brands and retailers can use today to drive them back. Like one of them that we first launched is the AI sales agent, which is the product that sits on the merchant's website that provides customers with conversational product discovery, personalized recommendations, and so on. And the goal is trying to recreate the experience of walking into a brick and mortar store and talking to an amazing salesperson. Can we bring that technology online with AI? And can that lead to more personalized experiences and higher conversions? So that was our first and our flagship product. And then recently, we launched another one called the AI Content Marketer. And the AI Content Marketer works with the AI sales agent, understands what users are talking about, and then based on those topics and queries, automatically generates hundreds of high performing landing pages for them. So today, for a marketing team, it might take several months to create this type of content, Whereas now you can have AI Content Marketer doing the job for you, and these pages can now start ranking on Google or AI search engines like Chagy PD and Perplexity. driving net new traffic for the brand, right? So the AI content market is basically building on top of the AI sales agent and then sort of like, you know, helping drive net new traffic. And our newest, you know, agent, the third one, is what we call the AI data scientist, which is again, you know, think of it as an employee of the brand, where anybody should be able to ask product or business metrics related questions directly on Slack, just like they would ask a human employee, right? And they would be able to get those answers. instantly basically bringing data and insights to their fingertips. Like today, I'm sure they have this data scattered across multiple dashboards. When they have to make a decision, then they have to ask somebody who's going to take multiple days to kind of bring the insights forward, right? What if that happened quickly? Will that lead to better decision making and faster decision making and give more opportunities for the brands to succeed? So as you can see, we are building out this platform for AI agents for different functions, but these are not standalone. Speaker 1 (04:14.44) agents, they're all operating on the same substrate of data and they're all collaborating with each other, right? And that to us is like the most distinguishing aspect of our platform. And tell me, so are you building your own economy? Great question. So we are compatible with all major e-commerce platforms. So whether the merchant is using Shopify, Salesforce, or even like the custom tech stack, right? We are a layer above that. you know, regardless of which e-commerce platform, we're not trying to replace, you know, any of the e-commerce platforms. You can think of it as an AI layer on top of that. So if any brand is thinking about how do I make my business AI ready? How do I do AI transformation for my organization? they should consider Big Sur AI as at least one of the candidates. I feel like most of the big e-commerce platforms are lagging behind in the AI market and especially in of the data. Speaker 2 (05:17.262) Like for example, Adobe actually had a product that they launched in 2018 or 19 called Adobe Sensei, but it's never really taken hold of the market within their own ecosystem. Do you feel as though you have an advantage over Speaker 1 (05:40.194) Yeah, mean, look, I think this is the nature of any startup, right? Which is when you're an incumbent, you have an existing product, you have an existing revenue stream that you also continue to to invest. We started at the beginning of the AI evolution, right? So we don't have any sort of baggage around incumbent products. So we are able to kind of build for the future. And what happens is, for all the incumbents, they certainly have the people and the resources to be able to do that, but they have to significantly disrupt their existing product roadmap and their ecosystem in order to kind of take that leap. So whereas we are basically bringing that leap and making it available for all brands and retailers. So that's why think for the first time in a decade or so since the mobile revolution, I feel there is a huge opportunity to reinvent the customer experience in e-commerce with AI, right? If you look at the last several years, the e-commerce user experience is somewhat stagnant, right? But for the first time, AI gives you that opportunity. So how could you unlock the power of the technology and take it forward without carrying the baggage of your existing products? Yeah, got it. What are you excited about here at Shop Time? So this is my third year at Sharp Talk in a row and you know two years ago when I came here we didn't have a product right like you know I came here to just to understand and explore you know. Speaker 1 (07:16.354) what people are talking about and whatnot. But most of our early customers have actually come from connections at ShopDoc. And so this time we have a bigger presence, we have more people from our team, we are talking at an event, and so we expect to make many more connections. That's basically the primary aspect of it. I consider ShopDoc as probably one of the best shopping retail events. I think there are a lot of like-minded retailers you know, we're here. And so we're trying to make that connection. The other thing that we have, we are announcing as part of the shop talk is our Big Sur AI innovators program. So where we are partnering with like forward thinking brands and retailers who really want to invest in sort of like shaping the future of shopping with AI, right? And so we work with partners who are in it for the long term, who are, you know, actually part of the transformation journey. So there are many benefits to this program that we offer, including like up to two million dollars for example in Big Sur driven sales completely free. you know when we partner with them as I said like you know we want to work with them for a long time and so that's one of the key programs that we're announcing as part of this so really excited lots of things going on. I should ask you this upfront, but what is your role with... I am the co-founder and CEO of the company. Yeah. Speaker 2 (08:39.022) and is there anything that you have that you can share? beyond. Speaker 1 (08:49.742) Yeah, I mean I think I would characterize it this way, right? We are still at this early stages of like, know, truly utilizing the power of AI. So there is a, the technology itself is moving pretty rapidly as we all know, right? And so if you fast forward like two years, three years from now, can you build an e-commerce experience where a consumer going to a brand's website, every pixel on the screen is being generated in real time and personalized for that individual. People have talked about hyper-personalization for a very long time, but they have only been able to do it incrementally because of the technology limitations. Now we have the opportunity to kind of reinvent that and actually make every pixel on the screen useful for you. And so we are kind of moving in that direction, and think in the course of time, a short period of time, I think you could build that where you truly are amazed at what a brand can offer to that individual knowing what they are looking for. And as Big Sur, how are you? Speaker 1 (10:01.922) Yeah, we do not build our own LLMs. For LLMs, we use multiple players, we have used LLMs at different parts in our technical stack. And for each one of them, we might use a different player, right? Because at the end of the day, it's all about quality, cost, and latency trade-off. So for that particular task, who can get the highest quality at the lowest latency and acceptable cost? And we build it in a way that we understand the technology is moving so rapidly. And within a few months, there might be another LLM that's even better than what we're using. And we're able to switch as quickly as possible so that we can offer the most cutting-edge technology to our customers. So last question, we're on latency. DeepSeq has, let's just say that they've explored a new way of being able to distribute the LLM. Speaker 2 (11:01.294) The pricing is very good. Have you... any those models that are outside of their glass. Yeah, so when the DeepSeek news came out, we started exploring it like the very first day. And like you said, there is always a trade-off in one direction or the other. Latency being very high is OK if you're going to use it for offline tasks. Not as a chat, for example, where latency is super important, but for offline tasks, you could use that. So we are always open to new models, and DeepSeek is a fantastic example of how they are disrupting the LLM ecosystem itself. And so all this to mean that I think the cost of intelligence is going down, right? It's trending towards zero. And I'll compare what the LLM costs were like 18 months ago versus now, it's I think down by like 90, 95%. And if you just like project that, it's pretty much going to zero. And so we are building for that future and we want to be open to like all models because innovation can happen everywhere. Speaker 1 (12:12.3) So. anything No, I think super excited about shop talk, you know great great chatting with you Brent