Smart AI Insight

Turn a sea of tickets into summarized intelligent data in one click.Smart AI Insights offers quick access to the context of Issues and Epics in Jira. Save time and boost collaboration with our AI app.

Comment

Inspiration

At TitanApps, we're focused on changing the way the companies are managed. The recent surge of interest in Large Language Models (LLMs) in the tech world grabbed our attention. As we delved deeper, we found ourselves actively using GPT models in our daily work, particularly in conjunction with Jira. It became evident that Jira holds a wealth of data about the organization, processes, and teams, motivating us to harness the power of AI to uncover these hidden gems and insights.

Introducing Smart AI Insight

Smart AI Insight is a tool for summarizing the content in Jira. It effortlessly converts Jira tickets into team updates and release notes. Users select the tickets they want in the summary, and the AI tool does the rest, empowering users to make informed decisions, streamline project communication, and foster team alignment.

Our Development Journey

Creating Smart AI Insight was marked by innovation and determination. We began with idea selection using the BRIDGeS framework, focusing on improving company management. Using AI for Jira data summarization was a simple yet impactful concept. We started with a basic Proof of Concept in Google Sheets for initial validation. Through user testing and iterative refinement, we improved the user experience. Choosing Atlassian's Forge platform streamlined development and showcased our problem-solving skills.

Overcoming Challenges

Our journey wasn't without challenges

  • UI Kit Selection: We faced the challenge of selecting the right UI kit. Custom UI emerged as the solution for delivering the dynamic interface we needed.

  • Prompt Engineering: Selecting the right prompt was a significant challenge. To simplify our journey, we conducted dedicated experiments to formalize the research process and choose the most effective prompt.

  • Request Time Limitations: We introduced a queuing system to retrieve responses and ensure that responses exceeding time constraints were not prematurely terminated.

  • Token Limitations in Chat GPT: Token limitations were addressed through an approximation technique, enabling effective text processing.

Celebrating Achievements

One of our most significant achievements was the team's rapid ideation and iteration. In a short time, we developed a fully functioning app with a validated concept, all within the uncharted territories of AI and Forge.

Valuable Lessons Learned

Our journey from ideation to MVP and hackathon submission for Smart AI Insight was filled with challenges, insights, and revelations. It emphasized the importance of proof of concept, early user testing, and prompt engineering. It also highlighted the significance of navigating limitations and choosing the right tools and platforms for project success.

The Future of Smart AI Insight

In the immediate future, we're dedicated to mastering issue summarization. Our plans include improving issue filtering, expanding model configuration options, and introducing preset summary formats. Our vision for Smart AI Insight extends beyond its current capabilities. We aim to transform it into a comprehensive Jira Assistant, capable of processing information from diverse sources and serving a multitude of needs and use cases. This transformation will involve tapping data from different sources and using it to generate valuable insights.

Built With

Share this project:

Updates