RetrospectAI

Inspiration

The inspiration behind RetrospectAI emerged from our own experiences with sprint retrospectives. While these meetings are critical for continuous improvement in Agile development, we found that they often lacked the objectivity needed to uncover the root causes of issues and foster meaningful change.

The traditional retrospective is often influenced by the emotions and perspectives of team members, which can obscure the real problems. We wanted to create a solution that transcended these limitations, providing a fresh, data-driven perspective.

What it does

RetrospectAI is an innovative JIRA app that reimagines your sprint retrospectives. It begins by collecting data about the specific sprint for which you wish to conduct a retrospective. Through seamless integration with JIRA APIs, it retrieves all available data related to the sprint and its associated issues. Leveraging this comprehensive dataset, RetrospectAI uses generative AI to get a comprehensive sprint retrospective report.

How we built it

In building this application, we used JIRA REST APIs and the Forge UI Kit 2. We leveraged Open AI Large Language Models to identify patterns, and suggest actionable improvements, ultimately leading to more productive and data-informed retrospectives.

We are getting detailed information about the sprint, issues in the sprint including but not limited to worklogs, comments and all the project roles (example: developers, QA etc) which is used by AI to analyze and provide the sprint retrospective report.

Challenges we ran into

The biggest challenge was finalizing our idea. We had a lot of ideas but solutions for most of them already existed. Another challenge was completing the project in a very short duration because of our schedules. Most notably, none of our team members were well-versed in the technology stack used in Forge, including JavaScript and UI development. This posed a significant hurdle as we delved into the project. We had to navigate unfamiliar tools, frameworks, and concepts, which came with a steep learning curve.

Accomplishments that we're proud of and What we learnt

Our journey in developing RetrospectAI, despite the initial challenges, has been marked by several significant accomplishments that we are truly proud of. Most notably, we overcame our lack of familiarity with the technology stack used in Forge. Through sheer determination and relentless effort, we not only adopted these new skills but excelled in them. This was not just a technological achievement, but a testament to our team's commitment to continuous improvement and adaptability.

In addition, our ability to solve complex problems and design effective solutions evolved significantly throughout this project. We learned to break down intricate issues into manageable components, allowing us to devise innovative and practical solutions. This not only benefited RetrospectAI but also honed our problem-solving and solution design skills, which will undoubtedly be valuable in future endeavors.

What's next for RetrospectAI

One of the points we need to improve on is the User Interface and saving the reports so that the user has access to them later on. Currently we are not saving the reports.

Furthermore, we are dedicated to refining the data collection process. By enhancing the data we gather, we aim to provide even more profound and actionable insights to our users. This means delving deeper into the metrics, incorporating new data sources, and utilizing advanced data analysis techniques. Our goal is to empower teams with a clearer, more comprehensive view of their sprints, enabling them to make smarter decisions and drive greater efficiency.

RetrospectAI is poised to continually evolve, adapt, and respond to the ever-changing needs of Agile teams. We are excited about the future, as it holds the promise of delivering an even more valuable and data-driven retrospective experience to our users.

Built With

• azure-app-service
• forge
• forgeui
• gpt-3.5
• javascript
• jira
• python