ArcGIS Enterprise Visualizer (EV)

Inspiration

It takes a lot of time and efforts for Esri customers to deploy, configure, and manage. There are at more than five endpoints to locate and debug when error happens. So the initial idea of ArcGIS Enterprise Visualizer (ArcGIS EV) is to monitor, and debug components of ArcGIS Enterprise within ONE SINGLE site, and provide users to fix the problems.

What it does

Monitor Enterprise components
Give Portal & Server Access
Suggest Fixes
View Live Logs
Track Component Connection Status
ArcGIS RAG with multi-tier Semantic Caching

How we built it

We developed a backend app using node.js with Express.js framework as an abstract singleton to monitor components' status and the connection status between them. And expose APIs from this structure and secure connection with ArcGIS OAuth authentication. The components contained are, Portal for ArcGIS, ArcGIS Server, Web Adaptors for both Portal and Server, and Data Stores. Then we developed the frontend app using Node.js and React framework to query information from the backend APIs. Plus we trained a RAG model as a chatbot for users to understand the problems deeper and thoroughly. The training datasets include Esri documentations, public Internet resources, and the error messages we collected.

Challenges we ran into

A full-stack web-application takes a lot time in integration debugging. We had a lot of issues when testing End-To-End featurees.

##Accomplishments that we're proud of

ArcGIS EV is something that Esri customers really need, we solved a real-world problem.
The completeness of this project is high during hackathon, there is no mock-up features. Every frontend widgets' status and data come from backend from real-time Enterprise status.
The chatbot we embedded in problem-solving can provide useful insights and solutions that traditional approaches cannot.

What we learned

Teamwork. Full-stack design, dev, testing, and debugging.

What's next for ArcGIS Enterprise Visualizer (EV)

For now, it's more focused on specific problem debugging and managing (for example, rasterAnalytics service tools related on Image Server), we hope we can expand the scope to broader services in ArcGIS Enterprise.

Built With

daisyui
esri
express.js
node.js
rag
react
tailwind
vite
yarn

Submitted to

Esri's Weekend of Innovation 2025
- Winner 2nd Place

Created by

I worked on the UI layout, logo, presentation design, and animations associated with the application. Addtionally I helped the team with enhancing the visual storytelling of the StoryMap and developed a brand system for the application itself.

Alexander Long
I am the developer. I was in charge of system design, front-end, back-end development, testing, and debugging. And also the deployment of the testing Enterprise environment.

laosong-2020 Song
I focused on enhancing the clarity and layout of our application. Furthermore, I utilize tools such as ArcGIS StoryMaps and Figma to craft a concise and compelling presentation.

Cael Pudenz
Built the entire RAG backend using FastAPI, Redis, and ChromaDB, with support for semantic and document caching, local LLM inference (DeepSeek + Granite via LM Studio), and a custom ESRI-focused vector knowledge base.
Also contributed to the responsive frontend with a glassmorphism UI, live markdown rendering, and accessibility features.
Ensured 100% local, privacy-first architecture with smart caching, performance optimizations, and secure deployment.

shlokjain Jain
I worked on highlighting user feedback to show the severity of the problem we are tackling and gathered metrics to show the improvement our solution offers. I facilitated team meetings, worked on the presentation to explain the ins and outs of our application, and was the designated speaker for the group.

Austin Laniyi

Updates

laosong-2020 Song started this project — Jul 27, 2025 02:39 PM EDT

Leave feedback in the comments!

Log in or sign up for Devpost to join the conversation.