Pinecone 2.0 — Edo Liberty, Pinecone

All right, All right, welcome everyone. welcome everyone. welcome everyone. I want to tell you a little bit about uh I want to tell you a little bit about uh I want to tell you a little bit about uh the new the new uh knowledge layer and the new the new uh knowledge layer and the new the new uh knowledge layer and I'll start with telling you a little bit I'll start with telling you a little bit I'll start with telling you a little bit about theory of mind. Uh how many of you about theory of mind. Uh how many of you about theory of mind. Uh how many of you uh know what theory of mind is? uh know what theory of mind is? uh know what theory of mind is? All right, so a few folks. It's a uh All right, so a few folks. It's a uh All right, so a few folks. It's a uh term in psychology about uh our capacity term in psychology about uh our capacity term in psychology about uh our capacity to understand uh and have a mental state to understand uh and have a mental state to understand uh and have a mental state or a model of a mental state for other or a model of a mental state for other or a model of a mental state for other people. What do they know? What do they people. What do they know? What do they people. What do they know? What do they believe in? How do they act? That's what believe in? How do they act? That's what believe in? How do they act? That's what allows us to socialize. That's what allows us to socialize. That's what allows us to socialize. That's what allows us to lie to each other and so allows us to lie to each other and so allows us to lie to each other and so on. Uh but it's a crucial developmental on. Uh but it's a crucial developmental on. Uh but it's a crucial developmental stage for humans. uh in a necessary stage for humans. uh in a necessary stage for humans. uh in a necessary function. I'll I'll sort of explain it function. I'll I'll sort of explain it function. I'll I'll sort of explain it in the best possible way with a funny in the best possible way with a funny in the best possible way with a funny story just to break the ice. Uh about 15 story just to break the ice. Uh about 15 story just to break the ice. Uh about 15 16 years ago I uh worked at a company 16 years ago I uh worked at a company 16 years ago I uh worked at a company called Yahoo called Yahoo called Yahoo um that had a product called Yahoo um that had a product called Yahoo um that had a product called Yahoo Answers which was sort of like a Q&A Answers which was sort of like a Q&A Answers which was sort of like a Q&A forum. Uh we used then AI which is the forum. Uh we used then AI which is the forum. Uh we used then AI which is the stone age of AI you know it's like uh stone age of AI you know it's like uh stone age of AI you know it's like uh NLP and other statistical models to NLP and other statistical models to NLP and other statistical models to answer questions. But the funny thing is answer questions. But the funny thing is answer questions. But the funny thing is people would come to uh Yahoo and ask people would come to uh Yahoo and ask people would come to uh Yahoo and ask questions like am I fat?

questions like am I fat? questions like am I fat? Which uh is funny because obviously uh Which uh is funny because obviously uh Which uh is funny because obviously uh you know Q&A forum would wouldn't have you know Q&A forum would wouldn't have you know Q&A forum would wouldn't have any clue whether you're you're too fat any clue whether you're you're too fat any clue whether you're you're too fat or not. Uh but this is a funny way to or not. Uh but this is a funny way to or not. Uh but this is a funny way to say yeah people have a very bad theory say yeah people have a very bad theory say yeah people have a very bad theory of mind on sort of what the products of mind on sort of what the products of mind on sort of what the products that they use what information they have that they use what information they have that they use what information they have and what they don't have. And I want to and what they don't have. And I want to and what they don't have. And I want to argue that in some sense we are making argue that in some sense we are making argue that in some sense we are making that mistake with AI today in a more that mistake with AI today in a more that mistake with AI today in a more subtle and less sort of like uh silly subtle and less sort of like uh silly subtle and less sort of like uh silly way but in a just as fundamental of a way but in a just as fundamental of a way but in a just as fundamental of a mismatch. And so I will categorize mismatch. And so I will categorize mismatch. And so I will categorize roughly three kinds of knowledge. I roughly three kinds of knowledge. I roughly three kinds of knowledge. I think there there are many more but you think there there are many more but you think there there are many more but you know in broad stro in broad broad know in broad stro in broad broad know in broad stro in broad broad strokes this is not a bad way to strokes this is not a bad way to strokes this is not a bad way to categorize. First of is general categorize. First of is general categorize. First of is general knowledge. knowledge. knowledge. Think about that is what's available in Think about that is what's available in Think about that is what's available in the public domain. Things that models the public domain. Things that models the public domain. Things that models are trained on and we expect an LLM to are trained on and we expect an LLM to are trained on and we expect an LLM to just know out the gate. Uh things about just know out the gate. Uh things about just know out the gate. Uh things about just you know improving your Rust code just you know improving your Rust code just you know improving your Rust code uh asking you like you know explaining uh asking you like you know explaining uh asking you like you know explaining stuff about the legal terms or uh sort stuff about the legal terms or uh sort stuff about the legal terms or uh sort of exploring use cases for for products.

of exploring use cases for for products. of exploring use cases for for products. Um we know how to do that. Models are Um we know how to do that. Models are Um we know how to do that. Models are really good at that. When we use our really good at that. When we use our really good at that. When we use our agents they have that information. there agents they have that information. there agents they have that information. there is specific knowledge [clears throat] is specific knowledge [clears throat] is specific knowledge [clears throat] the companies of course the models the companies of course the models the companies of course the models wouldn't have uh any clue about whether wouldn't have uh any clue about whether wouldn't have uh any clue about whether we've uh implemented something specific we've uh implemented something specific we've uh implemented something specific in our codebase what a contract of a in our codebase what a contract of a in our codebase what a contract of a specific customer has of course this is specific customer has of course this is specific customer has of course this is information that's private to the information that's private to the information that's private to the company uh but we know how to deal with company uh but we know how to deal with company uh but we know how to deal with that and rag and search and of course that and rag and search and of course that and rag and search and of course pine cone veto dbs and so on uh solved pine cone veto dbs and so on uh solved pine cone veto dbs and so on uh solved these problems a long time ago there's these problems a long time ago there's these problems a long time ago there's still a lot to go but we by and large still a lot to go but we by and large still a lot to go but we by and large know how to do it and then I would say know how to do it and then I would say know how to do it and then I would say there is tribal knowledge. Okay, tribal there is tribal knowledge. Okay, tribal there is tribal knowledge. Okay, tribal knowledge is the sort of general knowledge is the sort of general knowledge is the sort of general knowledge inside the company. This is knowledge inside the company. This is knowledge inside the company. This is the kind of information that uh an the kind of information that uh an the kind of information that uh an employ a seasoned employee has that a employ a seasoned employee has that a employ a seasoned employee has that a new hireer does not. Okay, it's roughly new hireer does not. Okay, it's roughly new hireer does not. Okay, it's roughly how we do things, what our processes how we do things, what our processes how we do things, what our processes are, who's in charge of what, uh what are, who's in charge of what, uh what are, who's in charge of what, uh what are the guiding principles, what are the are the guiding principles, what are the are the guiding principles, what are the you know sort of how we get stuff done you know sort of how we get stuff done you know sort of how we get stuff done in this company. Okay. um that is not a in this company. Okay. um that is not a in this company. Okay. um that is not a that's not general knowledge because that's not general knowledge because that's not general knowledge because it's it's specific to your company but it's it's specific to your company but it's it's specific to your company but it's also not a specific piece of it's also not a specific piece of it's also not a specific piece of knowledge because you can't just get it knowledge because you can't just get it knowledge because you can't just get it by searching. You can't just get it by by searching. You can't just get it by by searching. You can't just get it by plucking it off one place and say yes plucking it off one place and say yes plucking it off one place and say yes this document says uh that this is our this document says uh that this is our this document says uh that this is our culture. It's the sum total of a lot of culture. It's the sum total of a lot of culture. It's the sum total of a lot of uh things. Um and so if our theory of uh things. Um and so if our theory of uh things. Um and so if our theory of mind of most of our agents in big mind of most of our agents in big mind of most of our agents in big companies today should be something like

companies today should be something like companies today should be something like a new hire. They're brilliant. They have a new hire. They're brilliant. They have a new hire. They're brilliant. They have tools. They have we have very high tools. They have we have very high tools. They have we have very high expectations of them. But they are by expectations of them. But they are by expectations of them. But they are by and large clueless. Okay. They have no and large clueless. Okay. They have no and large clueless. Okay. They have no idea about your company's goals, idea about your company's goals, idea about your company's goals, culture, priorities, processes. Um they culture, priorities, processes. Um they culture, priorities, processes. Um they have to read a bunch of documents uh to have to read a bunch of documents uh to have to read a bunch of documents uh to be able to get any semblance of be able to get any semblance of be able to get any semblance of understanding of where they are in the understanding of where they are in the understanding of where they are in the world. Um and really they start most world. Um and really they start most world. Um and really they start most tasks from scratch. Okay. they really tasks from scratch. Okay. they really tasks from scratch. Okay. they really start kind of wake up and every time you start kind of wake up and every time you start kind of wake up and every time you start your your session, uh, they're start your your session, uh, they're start your your session, uh, they're like a new hireer on their first day on like a new hireer on their first day on like a new hireer on their first day on the job. the job. the job. So, I would argue that what they need is So, I would argue that what they need is So, I would argue that what they need is not necessarily to be smarter, to have not necessarily to be smarter, to have not necessarily to be smarter, to have better tools, but to have something new, better tools, but to have something new, better tools, but to have something new, which I would call a knowledge layer which I would call a knowledge layer which I would call a knowledge layer going forward. Um, and I have to explain going forward. Um, and I have to explain going forward. Um, and I have to explain to you what I think a knowledge layer to you what I think a knowledge layer to you what I think a knowledge layer has to actually do. But just I'll define has to actually do. But just I'll define has to actually do. But just I'll define it generally before explain to you what it generally before explain to you what it generally before explain to you what we do. we do. we do. Um Um Um so at the very least it needs to be so at the very least it needs to be so at the very least it needs to be persistent. We talked about this is the persistent. We talked about this is the persistent. We talked about this is the company knowledge, processes, company knowledge, processes, company knowledge, processes, capabilities and goals and so on. It has capabilities and goals and so on. It has capabilities and goals and so on. It has to be persistent. Uh it sure it uh feels to be persistent. Uh it sure it uh feels to be persistent. Uh it sure it uh feels like putting this thing together is like putting this thing together is like putting this thing together is going to be a monumental effort. So you going to be a monumental effort. So you going to be a monumental effort. So you want to do it once or at least very want to do it once or at least very want to do it once or at least very rarely. Uh and of course all your agents rarely. Uh and of course all your agents rarely. Uh and of course all your agents in company need to have access to this in company need to have access to this in company need to have access to this thing. The second thing is that you thing. The second thing is that you thing. The second thing is that you would argue that this [clears throat] would argue that this [clears throat] would argue that this [clears throat] knowledge layer needs to be uh highly knowledge layer needs to be uh highly knowledge layer needs to be uh highly specialized. If your agent uh is in

specialized. If your agent uh is in specialized. If your agent uh is in charge of HR practices or optimizing charge of HR practices or optimizing charge of HR practices or optimizing your uh your kernels in your codebase or your uh your kernels in your codebase or your uh your kernels in your codebase or uh resolving customer issues that they uh resolving customer issues that they uh resolving customer issues that they would care about different things they would care about different things they would care about different things they would want to you know address they would want to you know address they would want to you know address they would want to have different information would want to have different information would want to have different information available to them. Um, and I would argue available to them. Um, and I would argue available to them. Um, and I would argue irrelevant of of specifically how it's irrelevant of of specifically how it's irrelevant of of specifically how it's created, I would argue that it's a good created, I would argue that it's a good created, I would argue that it's a good practice to have the domain experts in practice to have the domain experts in practice to have the domain experts in the company own those and have being the company own those and have being the company own those and have being able to interact with them and make sure able to interact with them and make sure able to interact with them and make sure that they actually uh are aligned with that they actually uh are aligned with that they actually uh are aligned with what the company wants to do. what the company wants to do. what the company wants to do. And finally, which is the hardest thing And finally, which is the hardest thing And finally, which is the hardest thing is they need to stay up to date and they is they need to stay up to date and they is they need to stay up to date and they need to stay concurrent constantly need to stay concurrent constantly need to stay concurrent constantly because companies change and you need to because companies change and you need to because companies change and you need to make information make information make information uh uh deprecated. When it becomes uh uh deprecated. When it becomes uh uh deprecated. When it becomes deprecated, if the company replaces its deprecated, if the company replaces its deprecated, if the company replaces its CEO, the very next day, everybody in the CEO, the very next day, everybody in the CEO, the very next day, everybody in the company knows who the CEO is. Okay? Even company knows who the CEO is. Okay? Even company knows who the CEO is. Okay? Even though 9,000 documents going from, you though 9,000 documents going from, you though 9,000 documents going from, you know, a day before yesterday to uh the know, a day before yesterday to uh the know, a day before yesterday to uh the last 10 years say it's somebody else.

last 10 years say it's somebody else. last 10 years say it's somebody else. This has become common tribal knowledge This has become common tribal knowledge This has become common tribal knowledge immediately. And everything else, even immediately. And everything else, even immediately. And everything else, even you know, every rag search will tell you you know, every rag search will tell you you know, every rag search will tell you otherwise. This is now the overriding otherwise. This is now the overriding otherwise. This is now the overriding fact. Okay. fact. Okay. fact. Okay. Um I will argue before we move on that Um I will argue before we move on that Um I will argue before we move on that I'm not suggesting that this knowledge I'm not suggesting that this knowledge I'm not suggesting that this knowledge layer should replace everything else layer should replace everything else layer should replace everything else we're doing. Our agents are already very we're doing. Our agents are already very we're doing. Our agents are already very capable and should keep being very capable and should keep being very capable and should keep being very capable. Uh they have access to LLMs of capable. Uh they have access to LLMs of capable. Uh they have access to LLMs of course tooling local files search of course tooling local files search of course tooling local files search of course like rag vector search text course like rag vector search text course like rag vector search text search and so on. And I would argue that search and so on. And I would argue that search and so on. And I would argue that uh the knowledge layer needs to be a uh the knowledge layer needs to be a uh the knowledge layer needs to be a separate entity that all these agents separate entity that all these agents separate entity that all these agents have access to as as I explained why have access to as as I explained why have access to as as I explained why this is sort of a fundamentally this is sort of a fundamentally this is sort of a fundamentally different kind of data source. different kind of data source. different kind of data source. So I want to tell you more about Nexus. So I want to tell you more about Nexus. So I want to tell you more about Nexus. Um Um Um um Nexus is a product that we're uh uh um Nexus is a product that we're uh uh um Nexus is a product that we're uh uh of course it's a knowledge layer but I of course it's a knowledge layer but I of course it's a knowledge layer but I need I want to explain to you what it need I want to explain to you what it need I want to explain to you what it does. uh of course so the the four parts does. uh of course so the the four parts does. uh of course so the the four parts are roughly connectors what I would call are roughly connectors what I would call are roughly connectors what I would call context context context uh uh tasks and queries I'll explain uh uh tasks and queries I'll explain uh uh tasks and queries I'll explain each one of them uh now so connectors of each one of them uh now so connectors of each one of them uh now so connectors of course do the obvious things there's course do the obvious things there's course do the obvious things there's nothing much to talk about them um uh nothing much to talk about them um uh nothing much to talk about them um uh they bring data into the platform uh they bring data into the platform uh they bring data into the platform uh respect access uh rights and uh and

respect access uh rights and uh and respect access uh rights and uh and control and so on um very dull as a control and so on um very dull as a control and so on um very dull as a topic of discussion but incredibly topic of discussion but incredibly topic of discussion but incredibly important in enterprises, incredibly important in enterprises, incredibly important in enterprises, incredibly important to do right. Uh and everybody important to do right. Uh and everybody important to do right. Uh and everybody who's built that knows how painstaking who's built that knows how painstaking who's built that knows how painstaking it is to get this thing to work well. it is to get this thing to work well. it is to get this thing to work well. Um Um Um the interesting sort of the first the interesting sort of the first the interesting sort of the first properly interesting part is um the properly interesting part is um the properly interesting part is um the context. So the context think about that context. So the context think about that context. So the context think about that as the as the as the um the sum of assets that the that Nexus um the sum of assets that the that Nexus um the sum of assets that the that Nexus or this knowledge layer saves about a or this knowledge layer saves about a or this knowledge layer saves about a specific topic. Think about this as a specific topic. Think about this as a specific topic. Think about this as a context for HR, for operations, for context for HR, for operations, for context for HR, for operations, for engineering and so on. Okay, of course engineering and so on. Okay, of course engineering and so on. Okay, of course it has the sources, the connections to it has the sources, the connections to it has the sources, the connections to the data and so on. But then it has the the data and so on. But then it has the the data and so on. But then it has the manifest. Okay, the manifest is the manifest. Okay, the manifest is the manifest. Okay, the manifest is the first truly novel idea and and a first truly novel idea and and a first truly novel idea and and a important uh entity. Okay, important uh entity. Okay, important uh entity. Okay, the way that we today give context to the way that we today give context to the way that we today give context to our agents is with skills and plugins our agents is with skills and plugins our agents is with skills and plugins and uh commands and sort of markdown and uh commands and sort of markdown and uh commands and sort of markdown files we write ourselves.

files we write ourselves. files we write ourselves. In a knowledge layer, you don't do that In a knowledge layer, you don't do that In a knowledge layer, you don't do that anymore. Okay? You tell the you tell the anymore. Okay? You tell the you tell the anymore. Okay? You tell the you tell the knowledge layer, you tell Nexus what you knowledge layer, you tell Nexus what you knowledge layer, you tell Nexus what you care about in a manifest. You say what care about in a manifest. You say what care about in a manifest. You say what kind of tasks you're you want to kind of tasks you're you want to kind of tasks you're you want to complete, what entities you want to complete, what entities you want to complete, what entities you want to track, what kind of information you care track, what kind of information you care track, what kind of information you care about, and it's it's the knowledge about, and it's it's the knowledge about, and it's it's the knowledge layer's work to keep track of those layer's work to keep track of those layer's work to keep track of those files. Okay? And and organize its own files. Okay? And and organize its own files. Okay? And and organize its own data. data. data. Um Um Um uh and then it has the knowledge files uh and then it has the knowledge files uh and then it has the knowledge files themselves. So think about the manifest themselves. So think about the manifest themselves. So think about the manifest is the meta knowledge file or set of is the meta knowledge file or set of is the meta knowledge file or set of files. Um and then there's the knowledge files. Um and then there's the knowledge files. Um and then there's the knowledge the knowledge itself. uh think about the knowledge itself. uh think about the knowledge itself. uh think about this as five types of contents. One of this as five types of contents. One of this as five types of contents. One of them the the first kind is a semantic them the the first kind is a semantic them the the first kind is a semantic map. Okay, this is a file that organizes map. Okay, this is a file that organizes map. Okay, this is a file that organizes the catalog of where everything else is, the catalog of where everything else is, the catalog of where everything else is, the key schemas, the glosseries, all the the key schemas, the glosseries, all the the key schemas, the glosseries, all the information that you need to be able to information that you need to be able to information that you need to be able to operate this context.

operate this context. operate this context. Then there are of course a bunch of Then there are of course a bunch of Then there are of course a bunch of markdown files that contain information, markdown files that contain information, markdown files that contain information, decisions, memories, uh all sorts of decisions, memories, uh all sorts of decisions, memories, uh all sorts of unstructured information. If you've uh unstructured information. If you've uh unstructured information. If you've uh who here has heard about uh LLM wiki who here has heard about uh LLM wiki who here has heard about uh LLM wiki okay so LLM wiki is this uh idea that okay so LLM wiki is this uh idea that okay so LLM wiki is this uh idea that Andre Karpathy suggested to have agents Andre Karpathy suggested to have agents Andre Karpathy suggested to have agents basically basically basically manage some stash of markdown files to manage some stash of markdown files to manage some stash of markdown files to sort of keep their own notes of stuff. sort of keep their own notes of stuff. sort of keep their own notes of stuff. This is a very similar idea. This is a very similar idea. This is a very similar idea. Okay, sort of on steroids but similar Okay, sort of on steroids but similar Okay, sort of on steroids but similar concept but that's not enough. Context concept but that's not enough. Context concept but that's not enough. Context contains also a bunch of SQL tables for contains also a bunch of SQL tables for contains also a bunch of SQL tables for tabular data for aggregation for facts tabular data for aggregation for facts tabular data for aggregation for facts that they need to memorize. uh a vector that they need to memorize. uh a vector that they need to memorize. uh a vector database to be able to semantically database to be able to semantically database to be able to semantically search and text search and filter a search and text search and filter a search and text search and filter a bunch of information and all the bunch of information and all the bunch of information and all the information that it itself decides is information that it itself decides is information that it itself decides is necessary to keep and and remember and necessary to keep and and remember and necessary to keep and and remember and graph entities and graphics to remember graph entities and graphics to remember graph entities and graphics to remember uh again entities relationships uh uh again entities relationships uh uh again entities relationships uh causal chains and so on. Okay. And all causal chains and so on. Okay. And all causal chains and so on. Okay. And all of that is managed and created for you.

of that is managed and created for you. of that is managed and created for you. You don't touch any of this. You don't You don't touch any of this. You don't You don't touch any of this. You don't have to uh know how any of this is have to uh know how any of this is have to uh know how any of this is created. It's all maintained through the created. It's all maintained through the created. It's all maintained through the manifest. manifest. manifest. So these are this is the context. Now So these are this is the context. Now So these are this is the context. Now what are the tasks? What are the what are the tasks? What are the what are the tasks? What are the actions? These are the nouns. What are actions? These are the nouns. What are actions? These are the nouns. What are what are the verbs? What can we do with what are the verbs? What can we do with what are the verbs? What can we do with those uh objects? Uh first of all is of those uh objects? Uh first of all is of those uh objects? Uh first of all is of course import which is bring new course import which is bring new course import which is bring new information to the system, new files, information to the system, new files, information to the system, new files, new events, new uh concepts. new events, new uh concepts. new events, new uh concepts. Um curate is probably the most and most Um curate is probably the most and most Um curate is probably the most and most elaborate one of the two most elaborate elaborate one of the two most elaborate elaborate one of the two most elaborate functions. it's uh job is to take functions. it's uh job is to take functions. it's uh job is to take um to take the uh the say call it a file um to take the uh the say call it a file um to take the uh the say call it a file it doesn't have to be a file but let's it doesn't have to be a file but let's it doesn't have to be a file but let's say it's a new PDF or a new meeting say it's a new PDF or a new meeting say it's a new PDF or a new meeting transcription or a new uh presentation transcription or a new uh presentation transcription or a new uh presentation for the board and say now I need to for the board and say now I need to for the board and say now I need to update my view of the world okay so it update my view of the world okay so it update my view of the world okay so it takes that it takes the manifest which takes that it takes the manifest which takes that it takes the manifest which tells okay now what what things do I tells okay now what what things do I tells okay now what what things do I care about what concepts do I need to care about what concepts do I need to care about what concepts do I need to track what needs to be uh uh what do I track what needs to be uh uh what do I track what needs to be uh uh what do I need to memorize here and then it takes need to memorize here and then it takes need to memorize here and then it takes the semantic map that I told you the semantic map that I told you the semantic map that I told you contains all the like organizes where contains all the like organizes where contains all the like organizes where information is and understands okay fine information is and understands okay fine information is and understands okay fine I need to go update these tables these I need to go update these tables these I need to go update these tables these relationships my graph uh this uh these relationships my graph uh this uh these relationships my graph uh this uh these text indexes and so on um and when it's text indexes and so on um and when it's text indexes and so on um and when it's done it has essentially said okay fine I done it has essentially said okay fine I done it has essentially said okay fine I have assimulated the new information have assimulated the new information have assimulated the new information it's already it's it's sort of persisted it's already it's it's sort of persisted it's already it's it's sort of persisted in your context and we can move on Okay,

in your context and we can move on Okay, in your context and we can move on Okay, search is sort of a internal function search is sort of a internal function search is sort of a internal function very important. It basically abstracts very important. It basically abstracts very important. It basically abstracts with simple functions all the access to with simple functions all the access to with simple functions all the access to that information so that the agent that that information so that the agent that that information so that the agent that I'll tell you in a second doesn't have I'll tell you in a second doesn't have I'll tell you in a second doesn't have to know uh how to sort of access the to know uh how to sort of access the to know uh how to sort of access the internal parts of the context. internal parts of the context. internal parts of the context. Now comes uh the second I think most uh Now comes uh the second I think most uh Now comes uh the second I think most uh most uh interesting uh part of of the most uh interesting uh part of of the most uh interesting uh part of of the system which is noql. Noql is our way to system which is noql. Noql is our way to system which is noql. Noql is our way to to uh specify how we want information to uh specify how we want information to uh specify how we want information out of the context. It's not enough to out of the context. It's not enough to out of the context. It's not enough to just ask questions. If you're an agent, just ask questions. If you're an agent, just ask questions. If you're an agent, it's really important for you to give us it's really important for you to give us it's really important for you to give us the budget in tokens or in dollars. how the budget in tokens or in dollars. how the budget in tokens or in dollars. how much time you want us to invest in this much time you want us to invest in this much time you want us to invest in this sort of like do you want us to just a sort of like do you want us to just a sort of like do you want us to just a quick off-the- cuff kind of immediate quick off-the- cuff kind of immediate quick off-the- cuff kind of immediate answer or do you want us to go full tilt answer or do you want us to go full tilt answer or do you want us to go full tilt and scour everything we know to make and scour everything we know to make and scour everything we know to make sure we give you the most accurate sure we give you the most accurate sure we give you the most accurate answer uh and so on the structure and answer uh and so on the structure and answer uh and so on the structure and and so on. It's a you know it's an and so on. It's a you know it's an and so on. It's a you know it's an agentic tool. This is not like a chat agentic tool. This is not like a chat agentic tool. This is not like a chat interface. Okay.

interface. Okay. interface. Okay. Um and of course what you get back is Um and of course what you get back is Um and of course what you get back is like something that's agent friendly. like something that's agent friendly. like something that's agent friendly. This is uh you know grounded text and This is uh you know grounded text and This is uh you know grounded text and and kind of structured in the right way and kind of structured in the right way and kind of structured in the right way format. This again is not designed to be format. This again is not designed to be format. This again is not designed to be like a chat interface. This is really like a chat interface. This is really like a chat interface. This is really what your agents expect. Okay. what your agents expect. Okay. what your agents expect. Okay. Um so this is this is a uh um this is a Um so this is this is a uh um this is a Um so this is this is a uh um this is a uh trying to figure out how much time I uh trying to figure out how much time I uh trying to figure out how much time I have. have. have. Okay, I need to go a little bit faster. Okay, I need to go a little bit faster. Okay, I need to go a little bit faster. So So So uh this is a sophisticated crowd. I want uh this is a sophisticated crowd. I want uh this is a sophisticated crowd. I want to sort of uh pop the hood and show you to sort of uh pop the hood and show you to sort of uh pop the hood and show you a little bit about how things work. So a little bit about how things work. So a little bit about how things work. So uh if anything this is the interface I uh if anything this is the interface I uh if anything this is the interface I want to want to want to there are many many cool ideas uh that there are many many cool ideas uh that there are many many cool ideas uh that make this work but I want to tell you a make this work but I want to tell you a make this work but I want to tell you a little bit about just one of them. Okay. little bit about just one of them. Okay. little bit about just one of them. Okay. What you're seeing is the interface for What you're seeing is the interface for What you're seeing is the interface for Nexus. Okay. You will see at the top a Nexus. Okay. You will see at the top a Nexus. Okay. You will see at the top a question that the uh the agent issued question that the uh the agent issued question that the uh the agent issued and you see you see on the left uh the and you see you see on the left uh the and you see you see on the left uh the different steps that it went through.

different steps that it went through. different steps that it went through. Okay. What you see on the bottom right Okay. What you see on the bottom right Okay. What you see on the bottom right is probably the most interesting. You is probably the most interesting. You is probably the most interesting. You see generated code on the bottom right. see generated code on the bottom right. see generated code on the bottom right. What Nexus does which is very different What Nexus does which is very different What Nexus does which is very different than other systems. It's what we call a than other systems. It's what we call a than other systems. It's what we call a runtime coding agent. Okay. Unlike other runtime coding agent. Okay. Unlike other runtime coding agent. Okay. Unlike other coding systems in in that you're used to coding systems in in that you're used to coding systems in in that you're used to uh where the task is take a large code uh where the task is take a large code uh where the task is take a large code base and then help me edit it right but base and then help me edit it right but base and then help me edit it right but when the edits are done what I have is a when the edits are done what I have is a when the edits are done what I have is a piece of software and that is deployed piece of software and that is deployed piece of software and that is deployed that is the artifact okay what is that is the artifact okay what is that is the artifact okay what is running does not contain the model running does not contain the model running does not contain the model anymore it's just the code okay this is anymore it's just the code okay this is anymore it's just the code okay this is not what's happening here the in query not what's happening here the in query not what's happening here the in query time you should think about the engine time you should think about the engine time you should think about the engine essentially building something like a essentially building something like a essentially building something like a Jupyter notebook, something like uh like Jupyter notebook, something like uh like Jupyter notebook, something like uh like a Python ripple, right? And they a Python ripple, right? And they a Python ripple, right? And they literally write code and execute it and literally write code and execute it and literally write code and execute it and write code and execute and if they get write code and execute and if they get write code and execute and if they get the answer, they know what to do with the answer, they know what to do with the answer, they know what to do with it. And if it's, you know, if it's not it. And if it's, you know, if it's not it. And if it's, you know, if it's not what they expected, they rewrite that what they expected, they rewrite that what they expected, they rewrite that piece of the code and what you get in piece of the code and what you get in piece of the code and what you get in the end is a is a piece of code that the end is a is a piece of code that the end is a is a piece of code that completes the task that gets the completes the task that gets the completes the task that gets the information that you want to get. Okay, information that you want to get. Okay, information that you want to get. Okay, that is savable. that's rerunnable.

that is savable. that's rerunnable. that is savable. that's rerunnable. Okay. But it's also incredibly flexible Okay. But it's also incredibly flexible Okay. But it's also incredibly flexible because now the the answer is got not by because now the the answer is got not by because now the the answer is got not by by a query to a database. It's it's by a query to a database. It's it's by a query to a database. It's it's written by code which is of course written by code which is of course written by code which is of course incredibly flexible. Okay, that code is incredibly flexible. Okay, that code is incredibly flexible. Okay, that code is by the way, interestingly enough, if you by the way, interestingly enough, if you by the way, interestingly enough, if you replace if you write software this way, replace if you write software this way, replace if you write software this way, the the amount the amount of prompting the the amount the amount of prompting the the amount the amount of prompting that you need is significantly reduced. that you need is significantly reduced. that you need is significantly reduced. We went down from having something like We went down from having something like We went down from having something like 150,000 tokens to give our agents all 150,000 tokens to give our agents all 150,000 tokens to give our agents all the tooling they need to less than a the tooling they need to less than a the tooling they need to less than a thousand tokens to specify all the thousand tokens to specify all the thousand tokens to specify all the interfaces they need to get information interfaces they need to get information interfaces they need to get information out of Nexus. Okay. And so the the out of Nexus. Okay. And so the the out of Nexus. Okay. And so the the so what it looks like when you run it is so what it looks like when you run it is so what it looks like when you run it is like this. Okay. So on the left of like this. Okay. So on the left of like this. Okay. So on the left of course you'll see Nexus with uh NoQL course you'll see Nexus with uh NoQL course you'll see Nexus with uh NoQL answering the question and then on the answering the question and then on the answering the question and then on the right is sort the same uh task given to right is sort the same uh task given to right is sort the same uh task given to an agent with all the tooling to an agent with all the tooling to an agent with all the tooling to complete the task. Okay, but remember complete the task. Okay, but remember complete the task. Okay, but remember this on the right it's uh severely this on the right it's uh severely this on the right it's uh severely handicapped. It's like the employee on handicapped. It's like the employee on handicapped. It's like the employee on the first day on the job. They have to the first day on the job. They have to the first day on the job. They have to read a bunch. They have to figure out read a bunch. They have to figure out read a bunch. They have to figure out where the data is. They have to figure where the data is. They have to figure where the data is. They have to figure out what the task is. They have to write out what the task is. They have to write out what the task is. They have to write a bunch of code. Uh it's just it's just a bunch of code. Uh it's just it's just a bunch of code. Uh it's just it's just severely handicapped and there's severely handicapped and there's severely handicapped and there's I don't know if you okay stopped but um I don't know if you okay stopped but um I don't know if you okay stopped but um needless to say this is by far slower.

needless to say this is by far slower. needless to say this is by far slower. It's a lot more expensive in terms of It's a lot more expensive in terms of It's a lot more expensive in terms of token consumption. Uh but interestingly token consumption. Uh but interestingly token consumption. Uh but interestingly enough it's also a hell of a lot less enough it's also a hell of a lot less enough it's also a hell of a lot less accurate because again they don't have accurate because again they don't have accurate because again they don't have the right context, the right objectives the right context, the right objectives the right context, the right objectives and so on in your company. and so on in your company. and so on in your company. Um, I'll just say that [clears throat] Um, I'll just say that [clears throat] Um, I'll just say that [clears throat] we work uh obviously with a bunch of we work uh obviously with a bunch of we work uh obviously with a bunch of early access customers already. Uh, you early access customers already. Uh, you early access customers already. Uh, you see some results here, but I'll just see some results here, but I'll just see some results here, but I'll just sort of fly through them and say that sort of fly through them and say that sort of fly through them and say that across different domains, across across different domains, across across different domains, across different enterprises, across different different enterprises, across different different enterprises, across different kinds of task, you we universally see kinds of task, you we universally see kinds of task, you we universally see the same thing that if you shift to this the same thing that if you shift to this the same thing that if you shift to this paradigm, you get significant cost paradigm, you get significant cost paradigm, you get significant cost savings. uh 77 90 you know percent token savings. uh 77 90 you know percent token savings. uh 77 90 you know percent token consump token reduction is pretty consump token reduction is pretty consump token reduction is pretty common. Um common. Um common. Um uh it gets a hell of a lot faster uh it gets a hell of a lot faster uh it gets a hell of a lot faster anywhere from 20 30% to sometimes uh uh anywhere from 20 30% to sometimes uh uh anywhere from 20 30% to sometimes uh uh uh 77% faster. Okay. But and the most uh 77% faster. Okay. But and the most uh 77% faster. Okay. But and the most important thing is it actually becomes a important thing is it actually becomes a important thing is it actually becomes a lot more accurate as well in the same lot more accurate as well in the same lot more accurate as well in the same time. So this is really uh it's a slam time. So this is really uh it's a slam time. So this is really uh it's a slam dunk. I'll just wrap up by saying that dunk. I'll just wrap up by saying that dunk. I'll just wrap up by saying that uh um the that Nexus is coming out of uh um the that Nexus is coming out of uh um the that Nexus is coming out of early access and into public preview early access and into public preview early access and into public preview literally tomorrow. So literally tomorrow. So literally tomorrow. So go uh try it out.