Data engineers are constantly seeking tools that not only simplify their workflows but also enhance their ability to deliver robust data models to meet the demands of the business. Today, we're excited to announce the release of Cube Copilot, a groundbreaking assistant designed to empower data engineers to build and manage the universal semantic layer more efficiently than ever before.

## Empowering Data Engineers for Greater Productivity
Implementing a universal semantic layer is a foundational step in structuring and interpreting complex data. It bridges the gap between raw data and meaningful insights, enabling organizations to make informed decisions. However, building a semantic layer isn't always straightforward. It requires knowledge of various objects to construct and an understanding of the precise formats to declare them in.

With Cube Cloud, data engineers have access to a powerful open semantic layer platform that offers a plethora of features to model, aggregate, and serve data efficiently. But let's face it—keeping track of all the possible Cube objects and mastering the exact syntax of Cube's data modeling language can be challenging, especially for those who aren't working with it daily.

## The Challenge: Mastering Cube Modeling Language Without Daily Practice
Cube Modeling Language is a declarative, versatile language tailored for defining data models in Cube. It allows you to describe dimensions, measures, joins, segments, and more. While its flexibility is a strength, it also means there's a lot to learn and remember. For data engineers new to Cube or those who only dip into Cube Cloud occasionally, recalling every detail of the syntax or every available feature isn't always feasible.

Imagine you're trying to define a new measure or dimension but can't quite remember the exact syntax. Or perhaps you're aware that Cube supports a certain functionality, but you're unsure how to implement it in your data model. The natural solution would be to consult the documentation, but this interrupts your workflow and consumes valuable time.

## Help Is at Hand: Meet Cube Copilot
This is where Cube Copilot comes into play. We're introducing an intelligent assistant that's integrated directly into your Cube development environment. Cube Copilot is designed to help you edit and expand your data models faster.

Cube Copilot streamlines the coding process, allowing you to focus on designing effective data models. By offering contextually relevant suggestions, Cube Copilot helps minimize syntax errors and mismatches in your data models. Instead of searching through documentation, now you can spend more time building. 

For new users of Cube, Copilot accelerates the onboarding process. By providing guidance and suggestions, it helps you learn Cube organically as you work. Even experienced Cube users can benefit from Copilot by discovering new features or more efficient ways to implement functionality within their models.

### How Cube Copilot Enhances Your Workflow
As you work on your data model, Cube Copilot provides real-time suggestions based on the current context. Whether you're defining a new measure or adjusting a join, it offers relevant code snippets and syntax guidance to keep you moving forward smoothly.

Cube Copilot is aware of your database schema, including column types and existing tables. This means it can suggest appropriate dimensions and measures that align with your data's structure, reducing the likelihood of errors and enhancing model accuracy.

It doesn't just consider what's in the database; it also takes into account the existing parts of your data model. By understanding how you've structured your models so far, Cube Copilot can provide suggestions that are consistent with your current approach.

We've equipped Cube Copilot with numerous examples of how to build various Cube objects. This rich repository allows it to offer informed suggestions that reflect best practices and advanced techniques.

### Direct Prompting with Comments
One of the standout features of Cube Copilot is its ability to respond to natural language prompts. Simply write a comment in your code describing what you'd like to create, and Cube Copilot will generate the corresponding code for you. For example:

```
// Create a measure for total sales that sums the amount column
```

Upon entering this comment, Cube Copilot will provide the code necessary to define the `total_sales` measure, saving you time and reducing the need to look up syntax.

## Get Started with Cube Copilot Today
We're thrilled to bring Cube Copilot to our global community of data engineers. It's more than just an assistant; it's a partner that enhances your abilities and streamlines your workflow. Whether you're a seasoned Cube user or just starting your journey to unified data, Cube Copilot is here to support you every step of the way.

## Conclusion
Building a robust semantic layer is now more accessible than ever with the introduction of Cube Copilot. By simplifying the process of writing code and providing intelligent, context-aware assistance, we're confident that Cube Copilot will become an indispensable tool in your data engineering toolkit. Try Cube Copilot and experience a new level of productivity and efficiency in creating semantic layers. [Contact sales](https://cube.dev/contact?utm_source=cube&utm_medium=blog&utm_content=copilot%20blog) to learn more.


David Jayatillake 

All Things Cube

Introducing Cube Copilot: Your New Partner in Building Semantic Layers with Ease

We’re pleased to be featured in the July 10, 2024 issue of insideAI NEWS, formerly insideBIGDATA. The revamped publication offers comprehensive coverage of AI  technology and implementation strategies across multiple industries, including healthcare, security, financial services, automation, and agriculture. Below is a summary of our article, **“The Rise of Embedded Analytics and Embedded AI: Proportional Business Value at Last?” [Click here for the full text](https://insideainews.com/2024/07/10/the-rise-of-embedded-analytics-and-embedded-ai-proportional-business-value-at-last).**

Although business intelligence (BI) solutions have gone a long way to make more data available to non-technical users, challenges remain. Business users often give up when trying to find the right data because of the proliferation of data sources and dashboards. This is where curated data experiences — embedded analytics and embedded artificial intelligence (AI) can help — not just for customers and external use cases but in internal, user-facing applications, too.

## Access data where you need it with Embedded Analytics
With embedded analytics, users can access data when and where they need it most: in the context of their workflows. Instead of viewing a dashboard in the analytics platform and switching to the business application to act on it, businesses can curate the analytics experience within an application or a custom solution so there’s less jumping among applications and more immediate relevance. Taking it one step further, organizations can drive the experience using embedded AI so that users can bring up the right data within the business application using a simple voice command or chatbot.

But before organizations can deliver trusted embedded analytics and embedded AI, they need a universal semantic layer—an independent yet interoperable translation layer—between the data repository and the data-consuming endpoints. The semantic layer provides a consistent and trusted view of unified data by organizing, simplifying, and accelerating its consumption. Once a universal semantic layer is in place, curated data experiences are relatively easy to deliver, internally and externally. Although many organizations start with customer-facing applications, it is worth mentioning that substantial value can be derived quickly from focusing on internal processes first.

## AI and embedded analytics transform how data is used in an organization

There are many examples of internal processes that can benefit from embedded analytics and AI, including better employee experiences, decision-making in context, and better workflows and processes. When paired, a universal semantic layer and AI unlock profound capabilities in informing business users in real-time and context. For example, a universal semantic layer makes it possible to embed AI-assisted analytics into a tool like Salesforce, allowing for analysis on deals, prospects, and other key metrics to be done without context-switching — and via a near real-time process that can be as easy as querying an AI chatbot.

AI and embedded analytics powered by a semantic layer transform the way data is used in an organization, replacing the conventional, frequently fragmented method with one that is more integrated, perceptive, and useful.


Lissa Daniels

Artificial Intelligence

Data is the new oil, but it must be refined

## Discoverability and accessibility at the point of data consumption
### Why the Semantic Layer is the perfect place for a user-focused catalog

I’ve always thought the semantic layer is the perfect place for a catalog. Semantic layers are in production so their metadata remains fresh automatically as engineering work drives change. [Semantic layers](https://cube.dev/use-cases/semantic-layer) are at the point of consumption, so are the most relevant abstractions for data users to know about. Other data catalogs seem more geared towards data engineers who are wading through the swamp instead of business users who are looking to see what data they can get hold of to answer their questions.

Released today by Cube, [Semantic Catalog](https://cube.dev/product/semantic-catalog-and-ai-assistant) is centered around the semantic layer. The semantic layer contains the analytics vocabulary that your business users know about; the semantic layer is where data engineering meets business. This inherently makes this catalog more useful for business users. Yes, like other catalogs we have lineage from data source to consumer, but Cube’s Semantic Catalog contains the abstractions understood by business users: entities, metrics and dimensions. They can start at these abstractions, the analytics vocabulary they know about, and go backwards to understand their definition and forwards to understand where they are used in BI tools, data products, and elsewhere. This expands the use of the semantic layer to semantic intelligence.

This is not a data catalog, focused on a data engineer’s need to understand ELT, orchestration and data provenance. Semantic Catalog is focused on data analyst and business user needs, met by their data and analytics engineers.

![](https://ucarecdn.com/665e357f-dcde-4842-bbcd-df9aac427347/-/quality/smart/-/format/auto/)

## Allowing users to discover both data and metadata using AI

On top of the Semantic Catalog, we are also launching our new AI Assistant. This builds upon our launch of the AI API last month and further enhances with the metadata stored in Semantic Catalog. 

Business users can come here and ask questions in natural language. They don’t need to know any special syntax or terms. They don’t need to learn how to use a complex user interface. They can ask questions about what objects are in Semantic Catalog but also ask data questions and receive answers and results as line or bar charts here. They can ask these questions in the analytics vocabulary they already know. They are line-of-business experts; they don’t need to be part-time data analysts too.

This is not a BI tool. It is not our aim to build all the features required to deliver a modern BI experience - this is well served already by our partners. This is a data intelligence experience - a way to discover your data and use it in the same place with no technical knowledge required.

![](https://ucarecdn.com/c7838b9f-f9df-4234-83e5-e972dea4502f/-/quality/smart/-/format/auto/)

## Concluding Thoughts

With Semantic Catalog and AI Assistant, we have enabled users to have semantic intelligence - knowledge of how data flows through their stack into their semantic layer, what data is trusted and available in their Cube semantic layer, what downstream assets it relates to and also the ability to query it with natural language and AI.

Semantic Catalog and AI Assistant make Cube Cloud more useful to business users than ever before as well as our core engineering users.


Semantic Intelligence for All

The need for accurate, comprehensive, and easily accessible data is paramount. Ensuring that data is not only available but also meaningful and trustworthy is a challenge that many organizations face. 

Today, we are excited to announce the preview release of the [Semantic Catalog](https://cube.dev/product/semantic-catalog-and-ai-assistant), a new capability  designed to simplify data discovery and exploration. Let's dive into what Semantic Catalog offers and how it allows anyone to search and reuse connected data assets.

## What is Semantic Catalog?
Semantic Catalog is a groundbreaking innovation that provides a comprehensive, unified view of connected data assets. It is self-documenting and continuously up-to-date by default, based on the data models you define. This means that as you adjust and refine your data models within [Cube Cloud](https://cube.dev/product/why-cube-cloud), the Semantic Catalog automatically updates to reflect these changes, ensuring that your metadata is always up-to-date. This will be very different to the experience that Data Engineering teams have had previously with data catalogs, which require a great deal of manual entry and maintenance and invariably become stale. This is why we believe the semantic layer is the perfect place to have a catalog. Metadata from the semantic layer is in production and is kept current as a byproduct of engineering work.

## Key Capabilities of Semantic Catalog
### 1. Unified Search of Connected Data Assets
Previously, data engineering teams would look through the asset lists of multiple tools in their data stack to find specific objects. With Semantic Catalog, users gain the ability to search through a unified view of connected data assets seamlessly. 

Whether you are looking for modeled data in Cube Cloud, downstream BI content, or upstream tables, you can now find it all within a single, cohesive interface. This reduces the time spent jumping between different data sources and platforms, offering a more streamlined and efficient data discovery process for both engineers and consumers.

![](https://ucarecdn.com/b3a05158-db44-4593-9368-b1bbc3fc14fb/-/quality/smart/-/format/auto/)

### 2. Insights into Data Lineage and Relationships
Understanding the relationships and lineage of your data has often been a complex and time-consuming task. Semantic Catalog simplifies this by providing clear insights into the lineage, entities, relationships, dimensions, and metrics of your data. This is crucial for data governance, as it helps you ensure the quality and integrity of your data throughout its lifecycle. 

Additionally, this understanding empowers data analysts to make more informed decisions based on a comprehensive view of the data's history and interconnections.

__product screenshot: data graph from source tables to cubes and views__

### 3. Exploration of Downstream Content
Before Semantic Catalog, it was hard to know what the impact of engineering changes would be on downstream data assets like reports and dashboards. Engineers would have to check the code being run in these downstream tools to see what upstream assets they used.

Semantic Catalog facilitates the exploration of downstream charts and dashboards across all BI platforms. This means you can easily visualize how data flows from source tables through various transformations and into the reports and dashboards that stakeholders depend on. By visually navigating these connections using a [Data Graph](https://cube.dev/docs/product/workspace/data-model#data-graph), you can quickly identify the impact of engineering work to upstream data on downstream analytics, helping you to proactively manage and mitigate potential issues. For data analysts, starting with a Semantic Catalog search will help them find trusted data faster and also understand whether BI content already exists, rather than starting from scratch.

![](https://ucarecdn.com/5d4489af-9416-4c3f-9c17-fdb44663b177/-/quality/smart/-/format/auto/)

## Reducing Duplicate Efforts and Enhancing Collaboration
One of the standout benefits of the Semantic Catalog is its ability to reduce duplicate efforts in data modeling and analysis. In many organizations, data engineers and analysts often find themselves working in silos, leading to redundant work and inconsistencies in data interpretations. 

Semantic Catalog fosters better collaboration by providing a single source of truth that both data engineers and analysts can rely on. This unified approach not only increases productivity but also ensures that everyone in your organization is working with the same, trusted data.
Increasing the discoverability of data assets also reduces the likelihood of duplicate versions being created organically by different engineering teams or by analysts and engineers separately.

## Conclusion
The introduction of the Semantic Catalog marks a significant milestone in our ongoing commitment to empowering data-driven organizations. By providing a broad view of connected data assets, centered on the semantic layer, this product addresses some of the most pressing challenges in data discovery and management. From reducing duplicate efforts and enhancing collaboration to offering clear insights into data lineage and relationships, Semantic Catalog will greatly enhance the way your organization interacts with data.

We invite you to preview the Semantic Catalog and experience how it can improve data discovery. Stay tuned for more updates and enhancements as we continue to innovate and expand the preview capabilities. [Contact sales](https://cube.dev/contact?utm_source=blog&utm_content=Semantic%20Catalog%20blog) to learn more about Semantic Catalog and how it can connect your organization with trusted data.

Introducing Semantic Catalog (Preview)

Instant access to relevant and actionable insights is a game-changer. Recognizing this need, we are thrilled to introduce the latest addition to Cube’s expansive suite of tools: AI Assistant, now in preview. 

This new feature empowers non-technical users to ask questions in natural language and receive trusted answers based on your existing investment into Cube’s universal semantic layer. Let's delve into the transformative capabilities of [Cube's AI Assistant](https://cube.dev/product/semantic-catalog-and-ai-assistant) and how it can reshape your data accessibility and analytics processes.

## The Evolution of Data Accessibility
Traditionally, accessing and interpreting complex data has been a domain reserved for data analysts and engineers, which has led to data teams being overwhelmed with requests for data. With Cube's AI Assistant, we aim to remove these barriers and enable self-serve data access for all users–regardless of their technical expertise.

AI Assistant leverages LLMs to facilitate natural language querying, enabling users to gain quick insights by simply asking questions in their own words. A business query such as "What were the sales figures for Q2?" can be answered without writing SQL or navigating a technical user interface.

## Natural Language Processing for Seamless Interaction
Cube’s AI assistant does not generate SQL but instead generates an API request to the semantic layer - the text-to-semantic layer method not text-to-SQL. It does not rely on context from the semantic layer to enable it to write better SQL. It relies on the semantic layer to generate the correct SQL deterministically based upon the simple, constrained JSON format request. The AI Assistant generates this using retrieval augmented generation with metadata from the semantic layer and catalog. It requests known, understood and engineer-defined objects from the semantic layer. This results in greater consistency and transparency–[without hallucinations](https://cube.dev/blog/semantic-layer-the-backbone-of-ai-powered-data-experiences)–for both users and ease of debugging and validation for engineers.

![](https://ucarecdn.com/dc715311-3cf9-43f0-b894-452cf6fdeb72/-/quality/smart/-/format/auto/)

## Discover Modeled Data and Connect the Dots
The AI assistant can also answer metadata questions using the metadata stored in the Semantic Catalog. With this feature, you gain a holistic understanding of your data landscape. Users can easily discover available modeled data, downstream connected BI content, and upstream tables from the Semantic Catalog. This interconnected lineage provides deeper clarity and context, ensuring that all insights are framed within the broader operational data architecture of your enterprise.

![](https://ucarecdn.com/2c1a14ea-3675-4acc-a75c-103a63b97e28/-/quality/smart/-/format/auto/)

## Preview and Explore Interactive Charts 
Visual representation is invaluable in data analysis, transforming raw numbers into understandable trends and actionable insights. Cube’s AI Assistant offers users the capability to preview charts rapidly, transforming queried data into visual formats such as bar and line charts. This feature not only makes data more digestible but also enhances the ability to identify patterns and outliers at a glance.

For users who need deeper exploration, the AI Assistant seamlessly integrates with [Playground](https://cube.dev/docs/product/workspace/playground) 2.0. With a click of a button, users can easily transition from the conversational interface with a simple chart preview to an interactive exploration environment. This deep dive enables further interrogation of data, the overlay of different variables, and the customization of visualizations to extract even richer insights tailored to specific analytical needs. 

![](https://ucarecdn.com/5fc1d16f-1c6c-40bb-b6b0-0aefb8acf9a6/-/quality/smart/-/format/auto/)

## Democratizing Data Access within Organizations
Facilitating accurate, fast, and governed data is at the heart of Cube’s mission. By lowering the barrier for business users to work with enterprise data, the AI Assistant ensures that data teams aren’t overwhelmed by relatively simple ad-hoc data requests from stakeholders. At the same time, it leverages the data engineering work to build the semantic layer, which defines how the data model joins together and should be used. The meaning of data in your organization is still in the hands of your data engineers.

Stakeholders being able to safely, quickly and reliably self-serve with data—from marketing to finance to operations—leads to a more agile and responsive organization. Teams can align quickly on insights, pivot strategies based on real-time data, and enact changes with confidence, knowing they have the backing of reliable data.

[Contact sales](https://cube.dev/contact?utm_source=blog&utm_content=AI%20Assistant%20blog) to learn more about AI Assistant and how it can lower the barrier to entry for everyone in your organization.

Announcing Cube's AI Assistant: Empowering Every User with Data Intelligence

As data continues to grow in importance and complexity, Cube Cloud's latest innovations, built on the [universal semantic layer](https://cube.dev/use-cases/semantic-layer), ensure that enterprises can connect the entire organization with the right data. These advancements promise to simplify the way businesses discover, explore, and access modeled data, making it more accessible to everyone. 

With the introduction of [Semantic Catalog and AI Assistant](https://cube.dev/product/semantic-catalog-and-ai-assistant), Cube Cloud extends its utility even further to data analysts and business users within your organization. Let’s take a closer look at how Cube’s AI-powered universal semantic layer unlocks the value of trusted data.

## Semantic Catalog: Search a Unified View of Connected Data Assets
Semantic Catalog provides a comprehensive, unified view of connected data assets that is essentially self-documenting and always up-to-date, based on the data models you define. Users can effortlessly search through data modeled in Cube Cloud, explore downstream connected BI content, and examine upstream tables. 

This capability allows users to gain insights into the lineage, entities, relationships, dimensions, and metrics of their data. Additionally, users can visually navigate upstream connections to source tables and downstream connections to charts and dashboards across all BI platforms using a Data Graph. 

By providing a unified view, Semantic Catalog reduces duplicate efforts in data modeling and analysis, fostering better collaboration between data engineers and data analysts. This unified approach not only increases productivity but also takes the guesswork out of who to ask or where to go to find trusted data.

![](https://ucarecdn.com/051e290c-e12f-465e-af2a-2ee851462c13/-/quality/smart/-/format/auto/)

## AI Assistant: Ask Questions and Get Trusted Answers

Cube’s AI Assistant takes data accessibility to the next level. This feature allows users to interact with their data using conversational, natural language queries, providing trusted answers instantly. The GenAI-powered AI Assistant provides quick insights without technical expertise–no SQL code to write or drag-and-drop needed. 

Users can easily discover available modeled data, downstream connected BI content, and upstream tables from Semantic Catalog. Furthermore, AI Assistant allows users to preview charts and engage in deeper exploration in Playground 2.0 in one click. 

This feature lowers the barrier for business users to work with enterprise data by democratizing access further, enabling everyone in the organization, not just data engineers, to interpret and utilize data effectively, leading to faster and more informed decision-making across the board.

![](https://ucarecdn.com/5fc1d16f-1c6c-40bb-b6b0-0aefb8acf9a6/-/quality/smart/-/format/auto/)

## Cube’s Vision for Data-Driven Decision Making

For the first time, Cube's universal semantic layer is not just a tool for data engineers but an invaluable resource for everyone else in your organization. Administrators can define roles with specific permissions for Semantic Catalog, AI Assistant, and [Playground](https://cube.dev/docs/product/workspace/playground) access and apply those roles to users, creating a tailored user experience for non-data engineering users. 

Whether you are a data analyst searching for data to start your analysis or a business user seeking instant insights, the new capabilities in Cube Cloud ensure that everyone can participate in the data-driven decision-making process.
<Quote authorName="Artyom Keydonuv" authorAvatar="https://media.graphcms.com/mvIb32XHRoibCutwm7WA"  authorRole="CEO & Co-Founder, Cube"> As the demand for data grows, delivering value consistently, efficiently, and at scale becomes more challenging. Cube Cloud’s AI-driven capabilities enable more people throughout the organization to make more informed decisions and derive value from the single source of truth the universal semantic layer provides. When it comes to AI, the universal semantic layer is the critical piece of the puzzle to ensure accuracy without hallucinations. </Quote>

Cube is making it possible for a broader audience within organizations to benefit from data-driven insights. Data analysts can now interpret data more consistently, while business users gain a better understanding of the data, facilitating better, faster business decisions. Combining [AI with the universal semantic layer](https://cube.dev/blog/semantic-layer-the-backbone-of-ai-powered-data-experiences) to provide context, you can ensure accuracy and reduce hallucinations. 

## Transforming Data Interactions

While data continues to grow in importance and complexity, Cube Cloud's innovative features ensure that enterprises can unlock the full potential of their data assets and their people. By making modeled data easy to discover, explore, and access, Cube Cloud empowers organizations to deliver value to everyone from it. 

[Contact sales](https://cube.dev/contact "Contact sales") to explore the new Cube Cloud features today and transform the way your organization interacts with data.

Michael Hetrick

Unlock the Value of Trusted Data with Semantic Catalog and AI Assistant

At Cube, we’re on a mission to power the next-generation of data experiences by bringing consistency, context, and trust to all of your data. Today, we’re excited to unveil new advancements in Cube Cloud that accelerate the development of Generative AI (GenAI) and embedded analytics solutions. 

The latest updates include our new, native AI API, now in preview, and the general availability of Chart Prototyping. These innovations are built on the flexible foundation of Cube’s universal semantic layer, increasing the value you receive from your Cube Cloud deployment.

## Get Conversational with the AI API
The universal semantic layer is essential for building GenAI data applications. It provides context and governance to LLMs, reducing errors and hallucinations. Before the AI API, connecting GenAI to your data required either tedious manual work, such as vectorization and custom prompting, to provide the LLM with the right context and metric definitions, or going straight to a database without the governance and consistency of a semantic layer.

Customers like [spyne](https://cube.dev/case-studies/embedded-analytics-and-ai-chatbot-on-one-semantic-layer) and [Quantatec](https://cube.dev/case-studies/customizable-embedded-dashboards-and-natural-language-ai-queries) already benefit from the context provided by a universal semantic layer to LLMs. We were inspired by these customer use cases to simplify and accelerate development with the AI API. Now, Cube provides a standard interface for interacting with OpenAI as a turnkey solution for text-to-semantic layer queries using LLM-generated SQL. 

![](https://ucarecdn.com/e130a926-80c2-4679-9bf6-e3f05186e4f3/-/preview/1000x406/)

The new, native AI API reduces the complexity of typical application-LLM integrations. With a well-documented data model, shown below, and the AI API you can more easily create AI agents, chatbots, copilots, and Slack apps, using natural language interactions. Even better, your users will love it because it lowers the barrier to entry for working with data. The AI API will be generally available this fall.

![](https://ucarecdn.com/6dc9578a-9e83-4d6d-93f1-997a68cd5ed1/-/preview/932x295/)

## Streamline Development with Chart Prototyping
In addition to the AI API, we’re introducing Chart Prototyping for embedded analytics. Embedded developers and data engineers can now create and test charts within Cube Cloud’s Playground 2.0. This speeds up the development process, so that you can ensure insights are delivered in the most impactful way. 

Whether you’re integrating charts into internal tools or external customer applications, Chart Prototyping enables you to generate front-end TypeScript code, preview different chart types, and select your integration method via native REST or GraphQL APIs.

![](https://ucarecdn.com/1ca405f4-171b-4e94-a45f-2a7e9907999c/-/preview/1000x592/)

## Build Next-Generation Data Experiences
Cube Cloud’s new features are part of our commitment to helping you unlock the full potential of your data. With these capabilities, you can build secure, accurate, and cost-effective data experiences, enhancing the value of data for every user both inside and outside of your organization. 

[Contact sales to learn more](https://cube.dev/contact?utm_source=blog&utm_medium=site&UTM_Publisher=Cube&leads[…]tion&campaignid=701UH000009UsImYAK&utm_content=Contact%20Request) about how you can activate and leverage the AI API today. 

Introducing the AI API and Chart Prototyping in Cube Cloud 

Cube’s AI API is here, and it is available on Cube Cloud Premium and higher tiers. In our newest native API, we built the ideal architecture to extend the value of your Cube Cloud investment and accelerate the development of natural language capabilities with a turnkey solution, starting with OpenAI. 

The AI API includes Retrieval Augmented Generation (RAG) and prompt engineering, for an LLM to interface with the Cube’s universal semantic layer. This simplifies the development of AI applications and features in your data products.  

We leverage our deep knowledge of Cube’s universal semantic layer, its APIs, and metadata. You benefit from this API without needing to build your own RAG, prompt engineering, or vector store of your own. This saves a huge amount of time, effort, and cost compared to developing in-house. Now, you can focus on developing the best AI agents and chatbot experiences for your customers.

## Why Is Text-to-Semantic Layer Better?

The semantic layer is the foundation for trustworthy AI access to data using LLMs. With the public preview release of our new AI API you can now use Cube’s universal semantic layer with AI. The new AI API interfaces with our existing REST API to request the data the user has asked for; it does not generate raw SQL.

The semantic layer provides the ideal middleware to stand between LLMs and data. When asking an LLM to generate raw SQL on a database schema, it’s far too easy for it to hallucinate or get small things wrong as we saw in our webinar, [Delivering on the Promise of AI](https://cube.dev/events/delivering-on-the-promise-of-ai-increasing-accuracy-with-a-semantic-layer). 

The semantic layer provides context and also constrains what the LLM can answer with. LLMs have been trained on language itself–nouns, adjectives,  etc. Semantic layers provide LLMs with objects that are equivalent to these language elements: entities for nouns, dimensions/attributes for adjectives,  and measures for quantitative description. The LLM has the choice of the objects in the semantic layer and their attributes defined there, to answer questions with - not the whole knowledge it has been trained on.

This is just the first part, and is true of providing knowledge graph access to an LLM too. Semantic layers are special knowledge graphs, designed for querying data. Semantic layers, like Cube, provide APIs which have much more constrained input than raw SQL. If an LLM is forced to generate such an API request, it is much less likely to make a technically valid hallucination. Add to this, the fact that Cube is still compiling the request into raw SQL, doing the joins, aggregation, and filtering deterministically with our best-in-class compiler. The LLM is again constrained from work that it couldn’t reliably get correct.

![](https://ucarecdn.com/35b2c82f-5ba9-43c9-a13e-f470d8bc076d/-/quality/smart/-/format/auto/)

## Preparing Cube Data Models for Generative AI
You will use existing views within Cube Cloud to keep it DRY (Don’t Repeat Yourself); however, there are likely some modifications needed to work with the new AI API, as compared to other use cases.

Imagine that a data engineer from another company is reviewing your data model. The baseline test for preparing your data model for Generative AI is: **Is it clear enough that a data engineer from another company would understand it?** If the answer is yes because you’ve used clear, plain-language, readable terms and don’t have a lot of acronyms or company lingo, then it is likely that it can work with this API.

Don’t worry if this isn’t the case though, a few simple modifications can make all the difference. You don’t need to rework your whole semantic layer, just the parts that you want to expose to the AI API. First, ensure that you are working with Views in Cube Cloud, then use the following best practices to prepare it for use with the AI API:

- **Use clear, standard names.** Make names understandable, i.e. call a customer a Customer not some strange acronym like CustAcctNm that only people in your company use.
- **Add verbose descriptions.** Explain the member's purpose clearly. If you don’t want to put too much in the description field, in case it confuses human users, you can use a meta-tag to add more detail for the Cube AI API to pick up on. It’s helpful to mention in either if people use synonyms or acronyms to refer to that entity or member, if specific values for the field have special meanings etc. 
- **Make member names distinct.** If you have multiples of the same type of member, such as timestamp fields, try to make it very clear as to the use case for the field. For example, order_timestamp for when the order was made, and shipped_timestamp for when an order was shipped. Again, where there is ambiguity, you can describe how to treat it in the descriptions or meta-tag as if you were talking to a colleague.

![](https://ucarecdn.com/6dc9578a-9e83-4d6d-93f1-997a68cd5ed1/-/preview/932x295/)
## Testing the AI API with Postman
Once you have your views built and the AI API enabled (see the topic for the AI API in docs for more info on that), you’ll naturally want to try it out. An easy way to test is to use an API testing tool like Postman that lets you send API requests and inspect responses.

To get started, click the “Import” button in the top left or press Command+O. Then, paste the following example request in and hit enter:

```
curl \
 -X POST  \
 -H "Content-Type: application/json" \
 -H "Authorization: Bearer EXAMPLE-API-TOKEN" \
 --data '{ "messages": [{ "role": "user", "content": "what cities do we have the most sales in?" }]}' \
 https://YOUR_CUBE_API/cubejs-api/v1/ai/query/completions
 ```

This brings up a request. Replace YOUR_CUBE_API in the URL with your own Cube Cloud API, and EXAMPLE-API-TOKEN in the authorization header with a valid JWT for your Cube deployment.

In the request body, replace the question in “content” with something relevant to your own data set and the view(s) that you created.

After you hit Run, you will get a response that looks like the following: 
![](https://ucarecdn.com/50f63e71-5c35-41c9-ad1b-0bb93952cae3/-/preview/1000x561/)

This is a Cube query on your data, generated by AI! To run it, just copy the Cube query into a request to Cube’s REST API’s /load endpoint. Alternatively, you can add` "runQuery": true` to the AI API request body to run it inline.

<video width="100%" controls><source src="https://ucarecdn.com/16eee2fd-02ea-4311-90e7-ae3f4142c7c3/" type="video/mp4"/></video>

For example, in the video above, we ask the AI API for the cities where we have the most sales. It gives us back a Cube query using the Line Items view: total amount by city, top 10 descending. When we run the query, we can see that Los Angeles is where we have the most sales, followed by Seattle and Mountain View.

And now you’ve run your first Cube query using AI! Feel free to explore, add more messages to the conversation, and ask other questions to get a feel for what it can do.

## Get Started with the AI API
The release of Cube Cloud’s new AI API simplifies the development of Generative AI experiences for embedded developers and data engineers, while natural language interactions lower the barrier for every user of your product or solution–both inside or outside of your organization. 

We will continue to invest in the AI API and improve it over time, allowing all teams who deploy using our AI API to benefit from these enhancements as we release them. When it comes to how to use the Cube Cloud’s universal semantic layer with AI, we’ll do the heavy lifting, so you can focus on building sophisticated data experiences. To get started with the AI API in preview, [contact sales to learn more!](https://cube.dev/contact?utm_source=blog&utm_medium=site&UTM_Publisher=Cube&leads[…]tion&campaignid=701UH000009UsImYAK&utm_content=Contact%20Request)

A Practical Guide to Getting Started with Cube's AI API

I’ve been writing about the [semantic layer](https://davidsj.substack.com/p/semantic-layers-a-buyers-guide) for some time now, and regularly at that! I’ve said before that AI + SL is a match made in data, and I really believe that semantic layer adoption is necessary for any organisation that wants to use AI with their data - whether this is to make AI experiences for their customers or to democratise data access internally.

This is the basis on which Michael and I founded [Delphi Labs](https://www.delphihq.com/). We could see many startups being founded to offer this democratisation by using LLMs to write SQL queries, but we believed that this method was very limited. After experimenting with a few we realised that this method really suffers from hallucination as soon as a data model is more than a handful of tables. Delphi was the first company founded to focus on using third party semantic layers, like Cube, as the governed interface for LLMs to use data.

Through experimentation with all of the main semantic layer offerings, Michael and I found Cube. We found the [Cube Semantic Layer](https://cube.dev/use-cases/semantic-layer) to be the very best to use with LLMs and for our AI application in Delphi. We also met some great Cube folks and formed a strong partnership with them, leading to our announcement today.

Michael and I are joining Cube full time. I will be joining as VP of AI and Michael will be working alongside me to help bring AI offerings into Cube. We will be leveraging our knowledge of both AI and data engineering to enhance Cube’s product even further.

We see huge potential in joining Cube and real affinity with our mission. If you ask any senior data stakeholder, CXO/VP/HoX, whether they want to simply ask a question and get an answer back in seconds or minutes rather than hours, days or weeks… they say yes please! Ask the same stakeholders what a semantic layer is and why they should invest in one, and they require a lot of education and persuasion of the benefits. When it’s simply the infrastructure needed for their AI data experience to work, to be safe and to be reliable - then it’s like explaining why a train needs rails.

This move will help a much larger audience access what we’ve built and learnt, more easily. They can get the benefits of the best semantic layer out there and AI in one place. Get in touch to learn more about LLM and AI built on top of Cube’s semantic layer.

Michael and I are both experts in deploying and using semantic layers in practice, having both used them since leading data teams in our practitioner careers. We are excited to help contribute to Cube’s semantic layer progression that isn’t related to AI too! Given what I’ve written about before in my blog, it’s clear to all that I’m a semantic layer fan!

There will be more specifics to come about integrations, what we’re building at Cube, timelines, etc. as we move forward. 


The Delphi Labs team joins Cube.

Investing in Cube’s AI-driven future. 

Every business today is data-centric and the role of semantic layers for optimizing efficiency is undeniable. Serving as the bridge between raw data sources and end-point data sources, semantic layers offer a consistent, comprehensive perspective of data across the board. This helps businesses make decisions rooted and organized in accuracy and reliability. By centralizing the definition of semantics and metrics, data teams can eliminate inconsistencies and ensure a unified understanding across all data users, be in internal, external or AI bots. Let us look into how semantic layers streamline data management and analytics, simplifying the concepts of a semantic layer into digestible and relatable terms. 

## How a Semantic Layer Works in the Data Stack

[Semantic layers](https://cube.dev/use-cases/semantic-layer) are not a new concept, but their importance has grown with the increasing complexity of modern data stacks. Today, businesses use a variety of visualization tools, customer-facing analytics, [embedded analytics](https://cube.dev/use-cases/embedded-analytics), and AI agents. Each time a new data visualization or data app is implemented, metrics and/or logic need to be defined. If not managed properly, this can lead to inconsistencies and discrepancies and, ultimately wasted resources. Semantic layers address this issue by centralizing these definitions, ensuring consistency across different data consumption endpoints.

The semantic layer has been around for a while, embedded in Business Intelligence (BI) tools. However, the recent shift towards standalone semantic layers is driven by the need for a universal, easy-to-use, and highly discoverable data interface. The democratization of data has led to fragmentation and silos, impeding ease of use and leading to failed analytics projects. A standalone semantic layer addresses these issues, ensuring a single source of consistent data that is easier for data teams to work with. 

The implementation of semantic layers can significantly streamline data management processes. Sitting on top of [data warehouses](https://cube.dev/blog/getting-started-with-a-snowflake-semantic-layer), semantic layers provide data semantics (context) to various data applications. They work seamlessly with transformation tools, allowing businesses to define metrics, prepare data models, and expose them to different BI and analytics tools. This not only enhances efficiency but also ensures that all data consumption places are working with the same, accurate data. So, no matter if the data is being used by a person looking at a dashboard, or a LLM that is giving a person answers to questions, the data is consistent. All of this makes it easier for data teams to quickly deliver data for the various data consumers they work with internally and externally.

## Examples of how companies use semantic layers

Businesses like [Drift](https://www.drift.com/) and [Breakthrough](https://www.breakthroughfuel.com/) have successfully leveraged semantic layers to enhance their data management and analytics. Drift, a customer communication platform, uses semantic layers to ensure consistency in its metrics across different systems. This has not only improved their reporting capabilities but also enabled them to add new data in days instead of months. Breakthrough, a transportation management platform, uses semantic layers to provide consistent data across their SaaS and consulting platforms, significantly improving productivity.

Adopting a semantic layer in an organization is a step-by-step process. It begins with identifying inefficiencies and building a thoughtful semantic model. This model is then incrementally applied to different products, aligning metrics and data definitions. The process also involves identifying and fixing bugs in the old system, ensuring that the data is well-understood and well-documented. This is crucial for the success of analytics and AI projects.

### How AI agents, LLMs and the semantic layers work together

The semantic layer also plays a pivotal role in the realm of [generative AI](https://cube.dev/use-cases/llm-and-ai-semantic-layer). As AI agents are increasingly used for various applications, they need to access and analyze hard data. Due to their non-deterministic nature, AI agents are not good with hard numbers, they need to rely on the context training that has been done on the Large Language Model (LLM) they are fed from. Use of LLMs to generate code (like SQL) can be executed to get the hard numbers for the AI agent. This is where the semantic layer comes in, providing AI agents with the necessary data access and semantics via the LLM. 

However, the use of generative AI also presents new challenges. One of these is the need for a new interface for interacting with data. Another is the need for explainability, ensuring that users can trust the data and understand how the AI is working. The semantic layer can help address these challenges, providing a well-documented and well-understood data interface for AI agents. Semantic layers provide both valuable context and necessary constraint to empower retrieval augmented generation models. While much has been written about providing context via embeddings, we are now learning that constraining LLMs via a semantic layer - choosing from a defined universe of metrics - can be very useful for improving accuracy.  A recent test was performed with an AI Chatbot on data – with and without a semantic layer feeding the LLM. The results were dang clear – AI Chatbot with a semantic layer, outcomes were 100% accurate — AI Chatbot without a semantic layer, outcomes were 83% accurate.  

In summing up, semantic layers are an invaluable asset for businesses keen on optimizing their [data management](https://cube.dev/product/caching-and-data-performance) and analytics. They centralize metric definitions and semantics, ensuring consistency across various data points, which in turn, bolster efficiency and precision. The future will bring more data sources and more data applications, creating even more chaos in the data stack. The significance of semantic layers in navigating the intricacies of the contemporary data stack is poised to increase exponentially. The semantic layer plays a crucial role in increasing confidence in data for non-technical users, fostering the incorporation of analytics and AI projects. It also paves the way for generative AI to access and scrutinize data. Looking ahead, the impact of the semantic layer on data analytics and AI is expected to be even more profound.


Use Cases

Optimizing Data Management and Analytics Efficiency with Semantic Layers

Ever since the wide availability of ChatGPT in early 2023, Large Language Models (LLMs) have become a subject of intense interest and investment in nearly every industry. As we all try to wrap our heads around exactly how the power of generative AI will transform the world, here at Cube it was obvious to us that [semantic layers](https://cube.dev/use-cases/semantic-layer) would have a part to play. This would take shape as the integration of the Cube semantic layer into applications that retrieve information from data stores, perform analysis and return requested results to a user who is interacting with them using natural language, interpreted by an LLM. The exact return method of these results could be multimodal - natural language generation, tabular result sets, or even chained into visualization libraries to return dashboards, but the big idea is “ask a question, get an answer”. We weren’t sure exactly what it would look like, but we knew it was coming.

To prepare for this tectonic shift, we focused on learning about how builders were thinking about this problem and we released features that we felt would empower them such as our semantic document loader for LangChain and our expanded meta endpoints to make it easier to provide the useful context within Cube to the LLM, improving consistency and reducing hallucinations. We also engaged with many builders focusing on next generation analytics experiences driven by natural language query interfaces enabled by LLMs such as our partners at [Patterson Consulting](https://pattersonconsultingtn.com/) and [Delphi Labs](https://www.delphihq.com/) who have been instrumental in delivering AI-enabled analytics use cases for our customers. 

While we all felt that it made sense that an LLM based application would provide more accurate results when answering questions that required examining data within a database with the aid of well designed semantic layer, when compared with more simple zero-shot text-to-sql methods, we will be the first to tell you that we were long on customer anecdotes and light on empirical evidence. 

Recently a group of researchers from data.world (Juan F. Sequeda, Dean Allemang and Bryon Jacob) have published a [ground-breaking paper](https://arxiv.org/pdf/2311.07509.pdf) studying the accuracy of LLMs answering questions with a text-to-sql pattern and comparing those results with the accuracy of introducing a knowledge graph into the process. For those who may not be familiar, knowledge graphs are somewhat similar to semantic layers in that they are also an additional layer of business context that can inform usage of data. 

In addition to the results, the paper also puts forward the detailed methodology for evaluating the accuracy of LLMs in answering business questions asked of a database. They provide a schema, sample data, natural language questions and expected answers in a format that is fairly easy to replicate their experiment. Our partners at Delphi, who build an AI-enabled analytics application on top of semantic layers, quickly realized that this methodology could be adapted to perform a benchmark of the accuracy of their own solution. **[They have recently published some interesting findings of their own](https://delphihq.substack.com/p/delphi-at-100-video), that they were able to achieve an outstanding 100% accuracy with Delphi connected to a subset of the data.world test modeled in Cube as the semantic layer.** This is an incredible result and not only a validation of the ability of semantic layers to provide necessary context to improve LLM accuracy, but also speaks to their thoughtful prompt engineering and design choices. 

## Why are semantic layers the missing piece?

The two driving themes for why we need to use semantic layers with LLMs are context and constraint. We see more and more teams moving towards providing context through prompt engineering and doing things like providing the LLM with a knowledge graph to interface with. This is definitely a step in the right direction, but you also need to constrain their output.

Semantic layers provide both a knowledge graph and a constrained interface for an LLM. As a special type of knowledge graph, a semantic layer provides the LLM with a model of a world, composed of entities and their measures and dimensions. LLMs have been trained on language, and all language is composed of entities (nouns) and their dimensions and attributes (adjectives). Language also contains mathematical terms such as ‘total’, ‘running total’, ‘average’… as a result, LLMs are well-suited to knowledge graphs. 

The reason LLMs are also quite good at SQL is because of how many documents and articles (think Stack Overflow) contain SQL. However, SQL is a minuscule fraction of what LLMs have been trained on. An interface closer to natural language is a better fit for LLMs, and a constrained one, to reduce chance of error and hallucination, is even better. Requests to Cube’s REST API are both close to natural language (a requested list of metrics, dimensions and filters with few, if any, additional terms) and highly constrained (there is only really one way to form a request), if the LLM generates the request incorrectly this will almost always result in an error rather than a hallucination.

However, this has been theory so far, with some anecdotal evidence from using Delphi. What we needed was a benchmark, and now that the data.world one has been created - we have one.

## Testing Methodology
We undertook an ambitious benchmark test, following the footsteps of data.world and dbt Labs, who [tested their semantic layer on a subset of 11 questions](https://github.com/dbt-labs/semantic-layer-llm-benchmarking/tree/main) from the original benchmark. Our goal was to replicate dbt Labs’s methodology on our platform, utilizing Delphi in conjunction with the Cube semantic layer. **The outcome was remarkable - while dbt Labs achieved an impressive 83% across 8 questions (excluding 3 with multi-hop joins), Delphi + Cube reached a perfect score of 100% on all questions.**

The benchmark began by setting up a Snowflake data warehouse. We created the necessary tables from the data.world benchmark using the [DDL available in our GitHub repository](https://github.com/Delphi-Data/delphi-semantic-layer-llm-benchmark/blob/main/ACME_small.ddl).

Once the database structure was in place, we proceeded to populate it with the data files sourced from the insurance dataset [available on data.world’s GitHub](https://github.com/datadotworld/cwd-benchmark-data/tree/main/ACME_Insurance/data), originally developed by Object Management Group. This model, which involves a 3NF structure with multi-hop joins, was crucial in replicating the complexity of real-world data, setting the stage for an authentic benchmarking experience.

Next, we started building out the Cube semantic layer. This involved signing up for Cube’s free tier and creating a new project using the credentials from our Snowflake account. Using Cube Cloud’s built-in IDE, we developed the Cube schema available in our GitHub repository, including all tables and dimensions from the benchmark (even “extra” ones our questions did not use) and some basic measures.

Finally, we created a new free account on Delphi, connected Cube via its REST API integration, and then were off to the races. Working off the benchmark results sheet created by dbt Labs, we ran each question in a separate Delphi conversation and tracked the results. We replicated this 5 separate times to ensure that our results were consistent and not impacted by the non-determinism that’s often possible with LLMs.

Our results are publicly available, including the Cube queries Delphi generated. Delphi was able to answer 100% of the questions correctly, even including some where it was able to calculate the needed measures on the fly.

In line with our commitment to transparency and community collaboration, we have made our Cube schema and benchmark instructions publicly available. These resources can be accessed on our GitHub repository at [Delphi Semantic Layer LLM Benchmark](https://github.com/Delphi-Data/delphi-semantic-layer-llm-benchmark), inviting others to replicate our benchmark and share their findings.

## Conclusion

There is now empirical evidence that using a semantic layer with an LLM, for analytics, is superior to other methods. We have chained a number of proprietary methods in addition to this to deliver the best-in-class performance we see with Delphi. We invite everyone to take a look at the testing methodology above and take both [Cube](https://cubecloud.dev/auth) and [Delphi](https://auth.delphihq.com/en/signup) for a spin. We’d love to show you how easy it is to start asking questions of your data and feel confident that the answers your team will receive are grounded in the reality of your data.


Brian Bickell

Semantic layers provide LLMs context and constraint.

Semantic Layers are the missing piece for AI-Enabled Analytics

For many years already, we live in the data-driven world where accessing the data and deriving insights from it efficiently is paramount. This year, we experience an explosion of interest to artificial intelligence (AI) and large language models (LLMs) in particular, fueled by the latest developments in the technology and vast perspectives unfolding.

[LangChain](https://www.langchain.com), an extensive toolkit for working with LLMs, has become one of the most important building blocks for developers of AI experiences. At Cube, we know that our semantic layer is also an important building block for AI applications because Cube not only centralizes metrics calculations but also serves an [antidote to AI hallucinations](https://cube.dev/blog/semantic-layer-the-backbone-of-ai-powered-data-experiences).

## Semantic document loader

Today, we're happy to present Cube's integration with LangChain. It comes as the [document loader](https://python.langchain.com/docs/integrations/document_loaders/cube_semantic) that is intended to be used to populate a vector database with embeddings derived from the data model of the semantic layer. Later, this vector database can be queried to find best-matching entities of the semantic layer. This is useful to match free-form input, e.g., queries in a natural language, with the views and their members in the data model.

![diagram](https://ucarecdn.com/32e98c8b-a920-4620-a8d2-05d57618db8e/)

We're also providing an chat-based demo application (see [source code](https://github.com/cube-js/cube/tree/master/examples/langchain) on GitHub) with example [OpenAI prompts](https://github.com/cube-js/cube/blob/master/examples/langchain/utils.py#L48,L87) for constructing queries to Cube's [SQL API](https://cube.dev/docs/product/apis-integrations/sql-api). If you wish to create an [AI-powered conversational interface](https://cube.dev/blog/conversational-interface-for-semantic-layer) for the semantic layer, functioning similar to [Delphi](https://www.delphihq.com), these prompts can be a good starting point.

## Chat-based demo app, dissected

Here's what you can build with the all-new document loader for LangChain, a vector database, an LLM by [OpenAI](https://openai.com), and [Streamlit](https://streamlit.io) for the user interface:

<video width="100%" controls><source src="https://ucarecdn.com/5500be7a-ab78-4de8-abc9-02df575ad3e0/video.mp4" type="video/mp4"/></video>

See how the tables, columns, aggregations, and filters in the SQL query generated by an LLM match the human input. Check the [README file on GitHub](https://github.com/cube-js/cube/blob/master/examples/langchain/README.md) for pointers on running this demo application on your machine or skim through the following highlights.

**Ingesting metadata from Cube and populating the vector database.** The `ingest_cube_meta` function in the `ingest.py` file loads the data model from Cube using the all-new `CubeSemanticLoader`. Note that only [views](https://cube.dev/docs/product/data-modeling/reference/view) are loaded as they are considered the ["facade" of the data model](https://cube.dev/blog/complementing-data-graph-with-views). Loaded documents are then embedded and saved in the FAISS vector store, which is subsequently pickled for later use.

```python
def ingest_cube_meta():
    ...
    loader = CubeSemanticLoader(api_url, api_token)
    documents = loader.load()
    ...
    with open("vectorstore.pkl", "wb") as f:
        pickle.dump(vectorstore, f)
```

**LLM setup.** In the `main.py` file, dependencies are imported and the environment variables are loaded. The OpenAI model (`llm`) is initialized with the provided API key.

```python
import ...
load_dotenv()
llm = OpenAI(
    temperature=0, 
    openai_api_key=os.environ.get("OPENAI_API_KEY"),
    verbose=True
)
```

**User input and vector store initialization.** In the same file, Streamlit primitives are utilized to get user input:

```python
question = st.text_input(
    "Your question: ",
    placeholder="Ask me anything ...",
    key="input"
)
if st.button("Submit", type="primary"):
    check_input(question)
    vectorstore = init_vectorstore()
```

**Querying the vector store.** The vector store is queried for documents similar to the user's question. The best match's table name is extracted and taken as the best guess to try to create a prompt using the columns from vectorstore:

```python
docs = vectorstore.similarity_search(question)
    # take the first document as the best guess
    table_name = docs[0].metadata["table_name"]

    # Columns
    columns_question = "All available columns"
    column_docs = vectorstore.similarity_search(
        columns_question,
        filter=dict(table_name=table_name),
        k=15
    )
```

**Building the prompt and calling OpenAI.*** The OpenAI large language model is called with the constructed prompt, and the response is parsed to extract the SQL query and any associated filters:

```python
# Construct the prompt
prompt = CUBE_SQL_API_PROMPT.format(
    input_question=question,
    table_info=table_name,
    columns_info=columns,
    top_k=1000,
    no_answer_text=_NO_ANSWER_TEXT,
)
llm_answer = llm(prompt + PROMPT_POSTFIX)
```

![](https://ucarecdn.com/1252fc98-75db-4887-be82-227b60d477a9/)

## Wrapping up

Cube's integration with LangChain provides a seamless interface for querying data using natural language. By abstracting the complexities of SQL and leveraging the power of LLMs, it provides the builders of AI experiences with a user-friendly and error-prone approach to data access.

<Quote authorName="Josh Patterson" authorAvatar="https://ucarecdn.com/a3e94c6f-5d02-4b57-b523-bbfaf04f4890/8s5_8mFF_400x400.jpg" authorRole="CEO at Patterson Consulting">

It is essential for enterprises to leverage internal knowledge correctly with the power of LLM’s reasoning. Cube’s semantic layer integration with LangChain is a great example of how most product companies will eventually write smart integrations for LLMs to better power these reasoning engines.

</Quote>

Our partners at Patterson Consulting have been hard at work building with Cube, LangChain and LLMs. They have written a wonderful ebook as a primer to [Building Enterprise Applications with LLMs](https://pattersonconsultingtn.com/content/intro_to_llms_ebook.html).

**You can try the LangChain integration with a Cube Core or [Cube Cloud deployment](https://cubecloud.dev/auth/signup) today.** Join our Slack community at [slack.cube.dev](https://slack.cube.dev) to share your thoughts and opinions about this update.

Mikhail Nitsenko

Powerful tooling for building AI experiences with LLMs and the semantic layer

Introducing the LangChain integration

## The AI-powered data experience

We’ve long seen that the data experiences landscape has been very diverse. From dozens of general-purpose BIs to hundreds of vertical purpose-built analytical solutions, data consumers have many, many options to visualize and consume data.

The evolution of front-end technologies made it possible to embed quality analytics experiences directly into many software products, further accelerating the proliferation of data products and experiences. 

And now, with the arrival of large language models, we are living through another step change in technology that will enable many new features and even in the advent of an entirely new class of products across multiple use cases and domains—_including_ data. 

LLMs are taking the data consumption layer to the next level with **AI-powered data experiences** ranging from chatbots answering questions about your business data to AI agents making actions based on the signals and anomalies in data. 

![](https://ucarecdn.com/5c8bd295-4317-4820-beea-532f8c9ba391/)

## Semantic layer gives context to LLMs

LLMs are indeed a step change, but inevitably, as with every technology, it comes with its own limitations. **[LLMs hallucinate](https://en.wikipedia.org/wiki/Hallucination_(artificial_intelligence)); the [garbage in, garbage out](https://en.wikipedia.org/wiki/Garbage_in,_garbage_out) problem has never been more of a problem.** Let’s think about it like this: when it’s hard for humans to comprehend inconsistent and disorganized data, LLM will simply compound on that confusion to produce wrong answers.  

We can’t feed LLM with database schema and expect it to generate the correct SQL. To operate correctly and execute trustworthy actions, __it needs to have enough context and semantics__ about the data it consumes; it must understand the metrics, dimensions, entities, and relational aspects of the data by which it's powered. Basically—LLM needs a semantic layer.

The semantic layer organizes data into meaningful business definitions and then allows for querying these definitions—rather than querying the database directly. 

The ‘querying’ application is equally important as that of ‘definitions’ because it enforces LLM to query data through the semantic layer, ensuring the correctness of the queries and returned data. With that, the semantic layer becomes a solution for the LLM hallucination problem.

![](https://ucarecdn.com/c79d8e5a-ad67-43b3-ac4f-b3d323c0fb65/)

Moreover, the combination of LLMs and semantic layers can enable a new generation of AI-powered data experiences. At Cube, we’ve already witnessed many organizations build custom in-house LLM-powered applications, and startups, like [Delphi](https://cube.dev/blog/conversational-interface-for-semantic-layer), build out-of-the-box solutions on top of Cube’s semantic layer ([demo here](https://cube.dev/events/unleash-the-power-of-ai)). 

On the edge of this developmental forefront, we see Cube being an integral part of the modern AI tech stack as it sits on top of data warehouses, providing context to AI agents and acting as an interface to query data.

![](https://ucarecdn.com/ea4d11af-b3f9-43b9-b02e-b8fa8056c4b1/)

Cube’s data model provides structure and definitions that are used as a context for LLM to understand data and generate correct queries. LLM doesn’t need to navigate complex joins and metrics calculations because Cube abstracts those and provides a simple interface that operates on the business-level terminology, instead of SQL table and column names. This simplification helps LLM to be less error-prone and avoid hallucinations. 

For example, an AI-based application would first read Cube’s meta API endpoint, downloading all the definitions of the semantic layer and storing them as embeddings in a vector database. Later, when a user sends a query, these embeddings would be used in the prompt to LLM to provide additional context. LLM would then respond with a generated query to Cube, and the application would execute it. This process can be chained and repeated multiple times to answer complicated questions or create summary reports. 

### Performance

In regards to response times—when working on complicated queries and tasks, the AI system may need to query the semantic layer multiple times, applying different filters. 

So, to ensure reasonable performance, these queries must be cached and not always pushed down to the underlying data warehouses. Cube provides a relational cache engine to build pre-aggregations on top of raw data and implements aggregate awareness to route queries to these aggregates when possible.

### Security

And, finally, security and access control should never be an afterthought when building AI-based applications. As we mentioned above, generating raw SQL and executing it in a data warehouse may lead to wrong results. 

However, AI poses an additional risk here: since it cannot be controlled and may generate arbitrary SQL, direct access between AI and raw data stores can also be a significant security vulnerability. Instead, generating SQL through the semantic layer can ensure granular access control policies are in place. 

### And more...

We have a lot of exciting integrations with the AI ecosystem in store and can’t wait to share them with you. Meanwhile, if you are working on an AI-powered application with Cube – [let’s talk](https://cube.dev/contact)! 


Artyom Keydunov

—and the antidote to AI Hallucinations.

Semantic Layer: The Backbone of AI-powered Data Experiences

Today, we’re happy to present the newest addition to the plethora of [data tools](https://cube.dev/integrations) that work with Cube, the semantic layer for building data apps. [Delphi](https://www.delphihq.com) provides a way to interact with Cube in natural language and delivers insights straight to business users in Slack. 

![Diagram: Cube and Delphi](https://ucarecdn.com/ed0d51c4-2ff7-40cc-891f-32badcc57a51/)

Powered by the newest large language model (LLM) technology from OpenAI, Delphi brings the accessibility of self-service analytics to the next level and enables a compelling use case for semantic layers. See Cube and Delphi in action:

<iframe width="790" height="447" src="https://www.youtube.com/embed/FotEaaf20gY" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen allow="fullscreen;"></iframe>

Learn more about Delphi and [semantic superiority](https://davidsj.substack.com/p/semantic-superiority-part-4) from David Jayatillake, co-founder and CEO at Delphi Labs:

## Natural language interface for data

Our mission for Delphi is to solve the whole workflow, from question to answer, with data. Today, this is broken—self-serve BI tools haven’t delivered on this promise to date, and data teams are bombarded with ad hoc requests at a scale greater than ever before as businesses move to operate with data.

When we ask data leaders in the space what % of their organization is capable of self-serving with data, the answer is usually between 10 to 25%, with only 10% being very competent. It’s been three years since “dashboards are dead,” but unfortunately they’re still being used as the primary way to serve data. In my experience, no stakeholder ever wanted a dashboard. They wanted an answer, an explanation, a prediction, data to use elsewhere operationally, a graph or a slide to make a specific point.
This is the problem space we aim to solve with Delphi.

There are many other entrants who are trying to offer a natural language interface to data in this LLM gold rush. However, almost all of them are pointing the LLM at a database schema and getting it to generate SQL. We believe this is the wrong approach: every time a query is answered using this method, the LLM generates a minimal semantic layer on the fly in order to derive meaning from the data. As humans created the data structures and their meaning in the first place, why introduce a probabilistic element into the meaning of your data?

We believe that working with semantic layers curated by data folks is the best approach. It synthesizes the information already baked into the semantic layer, truly enabling safe and accurate answers. It also allows us to say that we can’t answer the question—a very normal outcome when asking a data team a question; LLMs connected directly to a data store will simply guess, as they don’t have a concept of a world and what does, and doesn’t, exist in it, which a semantic layer offers.

This is why we’re so excited to work with semantic layers like Cube—we’re already blown away by the interest we’ve had from the Cube community and hope to continue to work with you all!

As I’ve written in my [substack](https://davidsj.substack.com/archive), I believe the future of semantic layers is standalone. It’s what the people want.

## What’s next?

Here's what an early adopter of Delphi and Cube thinks:

> Getting started with Delphi on Cube was a breeze and we really quickly had a conversational AI interface to our data warehouse. The accuracy was great and some of the Delphi features really helped explain our data and ensure we knew how we were querying it. I'd highly recommend it for it's wow factor and enabling moving towards a data driven culture.

Interested? Implement your [semantic layer](https://cube.dev/use-cases/semantic-layer) with Cube by [getting started with Cube Cloud](https://cubecloud.dev/auth/signup) today or by [contacting us](https://cube.dev/contact). Also, [sign up for Delphi](https://docs.google.com/forms/d/e/1FAIpQLSc7nJc1Zeu5kEHcKrghbTPy3MSxJLOBsge0kj4Iiz5aqX6zJg/viewform) and see it in action at [slack.cube.dev](https://slack.cube.dev) in the `#random-delphi-demo` channel:

![Delphi in Slack](https://ucarecdn.com/ddf30e1e-c179-4fda-8aa8-2f75a05f808e/)
