Rovo Race Engineer

An AI pit-wall for Jira that surfaces the top race-critical risks in a sprint or release and tells teams the next best move.

Comment

🏎️ Rovo Race Engineer

Inspiration

In Formula 1, races are not won by drivers alone — they are won by race engineers. A race engineer continuously monitors telemetry, anticipates risks, and tells the team what to do next before it’s too late.

Software teams face the same problem every sprint and release: critical risks are visible in Jira, but buried under noise. Teams spend 30–60 minutes debating priorities, often missing the real blockers until it’s too late.

We wanted to build the AI pit-wall for software teams.

What it does

Rovo Race Engineer is an Atlassian Forge app designed to run as a Rovo agent inside Jira.

It analyzes Jira “telemetry” — issue status, priority, aging, and linked bugs/incidents — and answers one focused question:

“What are the top 3 race-critical risks that will block this sprint or release, and what should the team do next?”

Instead of generic summaries, Race Engineer returns:

  • A ranked top-3 risk list
  • Clear explanations of why each item is risky
  • Concrete recommended actions (“push”, “pit”, or “safety-car” decisions)

How we built it

The project is built entirely on Atlassian’s cloud-native stack:

  • Atlassian Forge (Node.js runtime) for secure app execution
  • Jira Cloud REST APIs to collect sprint and release telemetry
  • Rovo agent definitions to provide conversational access and reasoning
  • Structured action outputs so the agent remains deterministic and demo-safe

The app fetches issues using JQL, computes risk signals (priority, age, dependencies), and feeds a compact telemetry summary to the Rovo agent prompt, which produces ranked, actionable output.

Challenges

The biggest challenge was working with Rovo availability, which is currently rolling out and not enabled on all Atlassian Cloud tenants.

To address this:

  • The Forge app, agent definitions, and actions are fully implemented and deployable
  • The demo illustrates the exact behavior on a Rovo-enabled Jira environment
  • Clear installation instructions are provided so judges can run the agent on internal Rovo-enabled tenants

This allowed us to focus on idea quality, correctness, and real user value rather than environment limitations.

What we learned

  • Most delivery risk is already visible — teams just need it surfaced clearly
  • Atlassian Forge + Rovo is powerful when scoped to one high-impact decision
  • Judges (and users) value clarity and inevitability over feature breadth

Impact

Rovo Race Engineer helps:

  • Engineering leads de-risk releases faster
  • PMs and EMs focus on the right issues
  • Teams replace long risk debates with a clear, shared strategy

It’s a small app with a big shift in how teams think about delivery risk.

Built for Codegeist 2025 — Atlassian Forge + Rovo Challenge.

Built With

Share this project:

Updates