AI is being incorporated into nearly every facet of business; the possibility for enhanced productivity, efficiency, and efficacy grabs the attention of anyone attempting to become a competitive force in today’s high-demand market. The business of building data applications is no different than other enterprise areas seeking operational improvement through AI implementation.
Dan Zuckerberg, senior solutions architect at Databricks, and Jon Walls, account executive at Retool, joined DBTA’s webinar, Building Data Apps and AI-Powered Operations With Databricks and Retool, guiding viewers through the ways in which Databricks and Retool provides the means for data teams to rapidly develop powerful, AI-enabled data apps.
According to Zuckerberg, the challenges in delivering business outcomes with the modern data stack is entirely tied to complex, multi-stage data flows.
While orchestrating processes across all data, analytics, and AI use cases is critical, “more than 65% of organizations are using, at a minimum, 10 different data engineering and intelligence tools,” according to a 2020 IDC DataOps Survey.
Furthermore, siloed data teams, disconnected systems and proprietary data formats, and siloed tech stacks complicate the way enterprises keep up with the demands of modern business.
Zuckerberg then argued that the Databricks Lakehouse platform is ideal for harnessing a company’s data for analysis, unifying multiple personas—including BI and data warehousing, data engineering, data streaming, data science, and machine learning (ML)—and multiple data types in a single, secure platform.
Not only does it innovate in centralization, the Databricks Lakehouse also offers high performance. The platform is powered by Databricks SQL, a serverless data warehouse that runs SQL and BI applications at scale with up to 12x better price/performance, a unified governance model, open formats and APIs, and your tools of choice, without vendor lock-in.
The discussion then dove deeper into the ability to rapidly develop AI-enabled software, which, according to Walls, is a competitive advantage in today’s data-driven world. While 70% of top economic performers use software to differentiate themselves, challenges remain including:
- Inefficient operations
- Backlog of requests
- Inconsistent data
- Stale and insecure apps
Which is caused by:
- Explosion of SaaS apps
- Legacy architectures
- Resource constraints
- Heavyweight lifecycles
Ultimately, developing AI-enabled software takes time and resources, where making decisions on app frameworks, user interfaces, and services can slow an enterprise’s ability to ship out and update apps, Walls explained.
Retool provides a full business software stack in one place, enabling enterprises to build web or mobile apps, supported by backend logic and a storage layer, integrable with generative AI (genAI) services. Walls summed it concisely: Retool offers the ability to connect to proprietary data, build an interface on top of that data, and customize it further with code.
Taking the resources available—such as Databricks’ platform—Retool provides an environment to build applications on top of a layer that offers source control, authentication, authorization, and more.
For an in-depth discussion of how Databricks and Retool can be paired to accelerate software development in the enterprise, featuring demos, case studies, and more, you can view an archived version of the webinar here.