Complexity can be a significant detriment to any organization’s real-time data efforts. It’s a particularly big obstacle when building customer 360, requiring combining data from multiple systems, often in different formats with changing schemas. Then, there’s the lack of collaboration between different tools used for data processing, which creates friction between data producers and consumers and hinders an organization's ability to use its data assets fully.
Dataworkz aims to remove this friction and enable businesses to build customer 360 in minutes rather than months—at a fraction of the cost.
In a recent conversation with our newest Digital Champion, Nikhil Smotra, Chief Technology Officer and co-founder of Dataworkz, we explored Dataworkz's vision, the use of LLMs, and the advantages of Astra DB’s new vector search capabilities.
Empowering business users with an all-in-one approach
Dataworkz takes an “all-in-one” approach to combine data from various sources, apply complex transformations, and use a single platform for data processing. This solution caters to both technical and non-technical users, enabling the former to dive into the details and write SQL code while providing abstractions for non-technical users.
“The uniqueness of Dataworkz lies in its ability to empower business users to take ownership of their data and build a customer 360 in minutes,” Smotra said. “Traditionally, the data integration process could become a software integration nightmare, involving data engineers, data scientists, and various tools. Dataworkz streamlines this process by combining data transformations and AI into a seamless experience and providing an easy button for creating a customer 360.”
Harnessing the power of large language models
Large language models (LLMs) are pivotal in making this process more intelligent and efficient. Dataworkz uses LLMs in two primary areas. First, conversational data preparation enables users to describe their data preparation tasks in plain English, which are then converted into code by the language model. Users can interact with the system in real-time and make informed decisions about the next steps. Second, LLMs create question-and-answer systems and summarizations for proprietary customer data without coding. This abstraction of complexity lets business users focus on problem-solving rather than technical intricacies.
Smotra also emphasized the importance of vector databases in their workflow. “Vector search enables semantic search and contextual information retrieval, which enhances the customer experience and provides more accurate results,” he said. “By leveraging vector databases, Dataworkz can unlock transformative insights from private enterprise data, harnessing the power of LLMs.”
Collaboration between data producers and consumers is a critical aspect of data management. Dataworkz addresses this challenge through visibility and transparency. Activity streams powered by Astra DB notify stakeholders about creating, modifying, or deleting datasets, ensuring all stakeholders are aware of ongoing data ecosystem changes. Astra DB facilitates a shared understanding of data origins, enabling efficient use of an organization's data assets.
“The advantages of using Astra DB for Dataworkz include its collaboration capabilities, high-speed operation store, and recently introduced vector search,” Smotra said. “The collaboration between data producers and consumers is enhanced through Astra DB's ability to handle high-speed event generation, enabling real-time information retrieval. Vector search in Astra DB further strengthens Dataworkz's capabilities by providing fast lookups and efficient storage of vector embeddings.”
Looking ahead, Dataworkz aims to leverage LLMs to generate prescriptive and predictive analytical capabilities for its customers. These capabilities will enable the generation of previously inaccessible insights, empowering businesses to make data-driven decisions and drive innovation.
To learn more, watch Nikhil Smotra’s Digital Champion video here.
