Lessons Learned AI Agent

Lessons Learned AI Agent turns comments into stepping stones for continuous learning. Make your past work for you, let AI unlock the lessons that drive smarter decisions and growth

Comment

Inspiration

The inspiration for Lessons Learned AI Agent came directly from our own struggles with project retrospectives. As developers and consultants, we’ve all been part of retrospective meetings where valuable insights were discussed but often forgotten as time went by. We noticed a recurring pattern—lessons were learned, but rarely applied effectively in future projects. This led to the question:

How can we ensure that teams learn from the past and actually use that knowledge in real-time decision-making?

Jira, being our everyday tool, was the perfect platform to integrate such a solution. The idea was to build an AI-driven tool that not only captured lessons but also surfaced them at the right moment, empowering teams to make better, data-driven decisions.

What it does

It analyses all comments in a project taking relevant ones and transforming them into helpful and practical lessons learned items.

How we built it

The app is fully developed in Forge and utilises the Agent and Action modules.

Challenges we ran into

Integration Complexity:

Jira has a complex ecosystem, and ensuring our app worked seamlessly across different Jira products (Jira and Confluence) was a must.

Scaling and testing the App:

Trying to make Rovo connect different actions proved to be time consuming as different user inputs will get Rovo to come up with different outcomes.

Accomplishments that we're proud of

Creating something simple yet powerful that can be transformative for a team when utilized effectively.

What we learned

Throughout the journey of building Lessons Learned AI Agent, we gained insights that extended beyond the technical realm:

The Power of Context: Simply storing lessons is not enough. The true value lies in surfacing them at the right time, in the right context—whether it’s during sprint planning, issue resolution, or a post-mortem.

The Balance Between Automation and Control: While AI can be powerful, users still want a sense of control. Designing the system to augment—not replace—human decision-making was critical to its success.

What's next for Lessons learned AI Agent

Built With

Share this project:

Updates