Written by Johnson Ting, Software Consultant
Here at AC, we’re always looking for ways to stay ahead of the curve with emerging tools. Along with many of my teammates, I personally find the opportunities around agentic AI very exciting.
So, when the new Atlassian product Rovo came to the market, the first thing we thought about was, how can we utilise this to help our customers build on their project management experience?
With a next-gen AI tool integrated into your Atlassian stack, what can we do with it? How do we maximise its usefulness and effectiveness for our customers? What problems are we currently facing that can be easily solved with this new addition?
Instead of just reading the docs or running a standard internal demo, the best way to answer these questions is to bring the team together in a high-pressure, time-limited environment and roll up our sleeves to build something ourselves.
So we ran a Consultancy Rovo Hackathon.
ℹ️ Spotlight on Rovo
Rovo is Atlassian’s AI offering for search, chat, and agentic solutions. It helps teams find information and takes action across Atlassian products and connected third-party tools, enabling them work smarter with the information they already have.
Unlike a lot of AI assistants that feel like glorified search bars, Rovo is built to go deeper. It connects across Jira, Confluence, and your wider Atlassian ecosystem, meaning it can pull together context from tickets, documentation, project updates, and knowledge that’s usually scattered across multiple places.
So instead of spending time hunting through Jira issues, Confluence pages, and half-remembered notes, Rovo can help teams find what they need faster – and in many cases, summarise, interpret, and turn that information into something actually useful.
Have you enabled Rovo but aren’t realising value yet? AC’s Rovo AI Accelerator solution might be just what you need to increase adoption and maximise ROI.
The Challenge
At the start of our Rovo Hackathon, teams of 3–4 people were formed and given a simple brief:
Build a working proof of concept Rovo agent that solves a real consultancy or client pain point – by 5pm.
Any idea was allowed, as long as it was genuinely useful and realistic to build within a day.
By the end of the hackathon, each team presented:
- Their chosen use case
- The problem it solves
- How teams currently handle that problem
- How their Rovo agent improves the process
- A live demonstration of the POC
The goal wasn’t to create something perfect – it was to create something practical, compelling, and grounded in real-world scenarios.
And the results were impressive.
By the end of the day, teams (including mine!) had working agents that tackled real challenges across project delivery – and we came away with a much clearer understanding of where Rovo can add value for both consultancy teams and clients.
Which is better for building custom Rovo Agents – Studio or Forge? Discover which option is right for your needs with our Rovo Studio vs Atlassian Forge comparison.
What custom AI Rovo Agents did we build?
Here’s a flavour of what’s possible with Rovo agentic AI.
We built the following custom Rovo Agents:
- Workflow Helper Agent: Pulls in details about a specific work item and answers queries within the context of that item’s workflow – including where it sits now and what can happen next.
(It’s worth noting that this was my team… and I don’t want to brag, but our Agent won. Just for the record.) - Work Prioritisation Agent: When asked about the day’s priorities, this Agent analyses all of the user’s work and generates a clear, ranked list of what they should focus on today. Read the full use case here!
- QA Test Suite Creation Agent: This Agent pulls key details regarding acceptance criteria straight from Jira work items, and uses them to automatically build test cases in Zephyr Scale. Read the full use case here!
- AWS Incidents Insights Agent: When this Agent is called on a particular issue, the issue’s linked Jira asset and AWS asset are retrieved. The AWS logs are read so the situation is understood should the user wish to resolve the problem. Read the full use case here!
- Jira Sentiment Analysis Agent: This Agent reviews an issue’s comments and runs sentiment analysis to understand the tone. Based on that, it generates an appropriate response that can be added back as a new comment on the issue. If the Agent doesn’t detect clear sentiment in the comments, it simply stays quiet and doesn’t produce an output.
Do you have a recurrent pain point but don’t know how to solve it? Talk to us – the chances are we can create a custom agentic AI solution for you!
What did we learn (and what tips can I pass on to you?)
The hackathon really hammered home a few things for me:
- Rovo works best when it’s solving real problems.
The strongest ideas weren’t flashy AI experiments – they were rooted in the everyday frustrations we see in consultancy work. - Integration is everything.
Rovo starts to shine when it can pull from multiple places at once – Jira, Confluence, document stores, collaboration tools – and reason across all that context. - Speed unlocks creativity.
Working to a tight deadline forced all of us on the team to strip ideas back to core value. That pressure actually made it easier to spot what a strong, realistic agent looks like.
By actually building with Rovo, I got a much clearer sense of what it can do – and how to apply it in ways that genuinely help our clients.
Discover how Rovo can tackle your challenges with our tailored Rovo AI Accelerator solution – designed to unlock real-world value from agentic AI in your techstack!
Interested in exploring Rovo?
If you’re curious about how Rovo could support your teams – whether in delivery, pre-sales, onboarding, recruitment, or knowledge management – my colleagues and I would love to talk.
Get in touch to learn more about the Agents we built and how we can help you unlock the potential of Rovo.
