This is a submission for the DEV's Worldwide Show and Tell Challenge Presented by Mux
What I Built
I built a Jira app that helps development teams debug their delivery process. It combines Time in Status tracking with Flow Intelligence and AI-powered guidance to help teams understand where work gets stuck, predict when it will ship, and fix systematic bottlenecks.
Think of it as a debugger for your workflow, not your code.
My Pitch Video
Demo
Try it here:
https://marketplace.atlassian.com/apps/4028833452
The Story Behind It
Every developer has heard the question: "Why is this story taking soo long?"
The answer is usually buried somewhere in the process—work sitting in review, blocked dependencies, testing pile-ups.
Many teams see the same bottlenecks sprint after sprint, but without hard data, pushing for change is an uphill battle. So nothing changes.
I built Smart Guess because I wanted to bring the debugging mindset developers use for code to the way teams work. Instead of blaming people, you diagnose the system. The tagline says it all: Debug your process, not your team.
What makes it different from existing tools is the combination of:
Visibility — see exactly where time goes, right on the issue
Prediction — forecast when work will ship based on actual team history
Guidance — NoEsis, an AI assistant that helps interpret patterns and build the case for change
Technical Highlights
Backend:
- Atlassian Forge - Serverless cloud platform with FaaS runtime
- Node.js - Backend resolver logic
- Forge Storage API - Persistent data layer
- Jira REST API & Agile API - Sprint/board data integration
- OpenAI Assistants API - AI coaching (NoEsis) with thread-based conversations
Frontend:
- React - Component architecture with hooks
- Atlaskit Design System - Atlassian's official UI framework
- Vite - Modern build tooling and dev server
- @forge/bridge - Forge <-> UI communication layer
Top comments (0)