Every year at the Esri Developer & Technology Summit, thousands of GIS developers, administrators, and engineers gather in Palm Springs, California, to attend technical demonstrations, convene with other GIS professionals, and learn what’s fresh and upcoming in the world of ArcGIS. Every year, Esri developers and subject matter experts are on hand to offer technical support and guided learning. And every year, conference presenters and attendees encounter technological shifts that are more tectonic than the year before.
The 21st Esri Developer & Technology Summit was held March 10–13, 2026, and whether attending in person or watching videos of the Plenary Sessions, it was clear what was on everyone’s mind: enterprise GIS, low-code and no-code app building solutions, developer tools, and geospatial AI.
Nearly 4,000 attendees came together for this year’s summit, both in person and online. It isn’t difficult to imagine that all of them came away with a clearer understanding of how GIS is growing and changing, incorporating new enterprise workflows and AI models to streamline and enhance the work of GIS professionals.
“The purpose of this meeting is all about you,” said Esri founder and president Jack Dangermond to kick off the Plenary Sessions. “It’s you and your work, and growing and learning together.”
Integration and Configuration
Integration was one of the hottest topics at the two Plenary Sessions. So it was no surprise that Dr. Sud Menon, Esri’s corporate director of software product development, stressed the holistic nature of ArcGIS. It is an enterprise platform that does a lot more than just mapping—it connects data and people, operations and analytics. And web GIS and its architects are at the heart of the endeavor.
“A modern GIS is a web GIS,” said Menon. “It allows you to integrate all your sources and make them available anywhere and everywhere that you need them to be.”
Recent enhancements to ArcGIS—as well as presentations throughout the Plenary Sessions—demonstrate this kind of web-based systems thinking, with an emphasis on enterprise GIS and low-code/no-code app building solutions.
The new unified metadata editor and redesigned ModelBuilder extension in ArcGIS Pro, for instance, are all about streamlining the user experience. ArcGIS Enterprise, which allows organizations to run web GIS on infrastructure they control, features updates including new cloud storage options and the addition of ArcGIS Data Pipelines, a no-code data integration capability that is now available in ArcGIS Enterprise.
Jeremy Bartley, Esri’s chief technology officer for web mapping and geospatial web app technology, used a web map created with New York City’s 311 service request data as an example of how hosted feature layers operate within ArcGIS Online. It was an opportunity to display the most up-to-date versions of data visualization capabilities such as pop-ups and charts, and to demonstrate how feature layers can be shared through low-code/no-code software like ArcGIS Dashboards and ArcGIS Instant Apps.
Presentations from the City of Raleigh and Langan Engineering highlighted the idea that integration and configuration are at the core of modern GIS. In Raleigh, configurable apps, digital twins, and AI-assisted traffic analysis are used in tandem to support municipal operations and growth. The combination of custom apps, created with tools like ArcGIS Maps SDK for JavaScript, that operate within an enterprise GIS and configurable solutions with a low barrier of entry exemplifies the kind of integrated thinking that is becoming more necessary as GIS evolves and expands.
As these integrations become more customized and elaborate, developers can also extend ArcGIS to accommodate their organizations’ needs. This includes the use of ArcGIS Enterprise service interceptors that allow developers to inspect and modify REST-based service requests, as explained by Esri product engineer Pankaj Chaudhari. Service interceptors let organizations synchronize other software ecosystems, such as Salesforce, with ArcGIS. ArcGIS can also be extended through scalable, cloud-based environments like ArcGIS Enterprise on Kubernetes, which increase performance and reduce downtime for crucial apps.
Low-Code, No-Code . . . More Code
Low-code and no-code solutions have become central to the business of GIS, whether for enterprise workflows, public engagement, or field operations. These tools can expand the reach of ArcGIS and shorten the path from data to outcome.
“Tools like [ArcGIS] Experience Builder and Instant Apps have allowed us to put app creation in the hands of non-developers,” said Justin Greco, geospatial developer for the City of Raleigh. “And demand for web apps has never been greater.”
Dave Nyenhuis, an Esri senior product engineer, highlighted updates to ArcGIS Dashboards, including improvements to the theme panel and date selector, as well as a new data sources panel that simplifies the process of updating or repairing a dashboard. Instant Apps has also received some major enhancements. Esri product engineer Beth Romero detailed a workflow involving the Reporter and Manager templates, showcasing the AI-powered image extraction capability in beta within the Reporter template. This workflow also includes a new Web Editor template, which is designed for precise edits.
Within Experience Builder, updates such as a new login widget and dynamic Arcade expressions provide greater flexibility within a low-code solution. Esri software product engineer Wei Ying demonstrated the power of the new Arcade assistant, which can automatically generate Arcade expressions. Esri developer team lead Gavin Rehkemper showed off some of the capabilities of Experience Builder developer edition, including the ability to create custom widgets using map components within ArcGIS Maps SDK for JavaScript.
JavaScript Maps SDK has also seen many recent improvements over the previous year, according to Bjorn Svensson, group project engineer lead at Esri. For apps that require a greater degree of customizability than low-code/no-code tools provide, developer tools like JavaScript Maps SDK or ArcGIS Maps SDKs for Native Apps are the solution.
“There’s so much that’s been released in the last year,” said Svensson. “Batch editing for maximum productivity, track rendering to visualize real-time data, 3D flow render, 3D basemaps, calendar heat charts. With just a few lines of code, using the SDK’s components, you can build compelling and amazing experiences and make your data come alive with dynamic visualizations and client-side analysis.”
Strong support for web components, Esri’s Calcite Design System, and open-source libraries signals that the ArcGIS environment is becoming more component-based, and thus increasingly flexible for developers.
A Geospatial Platform Enhanced with AI
AI has, in many ways, been part of ArcGIS for a long time. But with the rise of generative AI tools, the way AI is incorporated into the platform is changing. The central thrust of this year’s Esri Developer & Technology Summit was this: AI is no longer a technology on the horizon. It now plays a central role in enhancing the ArcGIS ecosystem.
“Our vision in this space is a transformational one,” said Esri chief technology officer Jay Theodore. “Everything that we build is well-architected and well-integrated, because that’s fundamental to how AI gets infused in our GIS workflows.”
Theodore and other presenters at the Esri Developer & Technology Summit explained AI within ArcGIS as falling into three categories: AI tools and models, AI assistants, and agentic AI. Each has the potential to transform how GIS professionals interact with geospatial data.
-
Esri developers offered technical support and guided learning throughout the conference.
AI tools and models comprise powerful spatial and predictive analytics capabilities embedded within the software. Rohit Singh, director of Esri’s R&D Center in New Delhi, India, demonstrated new imagery and remote sensing models that can perform accurate damage assessments, road condition classification, and other analyses using multiple kinds of imagery. There are now more than 100 pretrained AI models available in ArcGIS Living Atlas of the World.
ArcGIS also provides tools to create and customize AI models. The Process Text Using AI Model tool within ArcGIS Pro, for instance, can extract data from unstructured text and turn it into spatial layers.
AI assistants, which enhance many of the tools and capabilities within ArcGIS, operate through natural language interactions. These assistants have a major role to play in the future of ArcGIS, a fact that was not lost on attendees. Besides the Plenary Sessions and the Keynote, a technical session on the fundamentals of AI assistants was by far the conference’s most popular offering.
ArcGIS Pro now includes an in-app assistant that uses natural language to help users explore documentation and create expressions and queries. ArcGIS Pro SDK can extend this assistant for custom workflows. ArcGIS Notebooks includes an AI assistant as well. It can explain code, troubleshoot errors, generate Python scripts, and help analysts work more quickly with notebook-based GIS workflows.
Agentic AI in ArcGIS enables systems to carry out tasks using geospatial data and tools, helping automate workflows and bring geospatial intelligence into broader agentic processes. With AI components in ArcGIS Maps SDKs, developers can build agentic mapping apps powered by either custom or out-of-the-box geospatial agents.
“This combination of AI with GIS is very powerful,” said Singh. “Not only can you use it to let AI take care of the drudgery of feature extraction, but [you can] also use it to solve some of the most complex challenges facing the planet.”