4 min read
AI Agent Studio
Enabling users to create context for different workflows.
My Role
Product Designer
Timeline
Hackathon, 24 hours
My Team
2 Product Managers
2 Engineers
1 Designer
What's the background?
Building a context driven agent for Atlassian's AI system (Rovo)
During an internal Atlassian hackathon, I explored how people could organize AI-assisted work around real projects, giving agents and chats shared context so teams could move faster across workflows like research, planning, and operations.
Our Approach
Creating custom agents based on chat history
Since this was a hackathon project, we had a very short amount of time to create our vision for how users would be able to create agents trained on their context from chat history. We decided to build the feature into the existing Rovo chat interface to increase visibility for the new experience.
Learnings
Key Takeaways
Participating in a hackathon with the other interns was a fun and memorable experience, and working closely together for 24 hours showed just how much you can get done in a short span of time.
- The design vision doesn't need to be perfect — Especially in a short timeframe like a hackathon, it's important to align with Engineering on how the experience should look and function. At the same time, overcomplicating the design can be detrimental because Engineering has limited time to build.
- It's important to understand the business value when building — We explored many different ideas for the project, and although several were exciting from a design and engineering perspective, they didn't make sense from a business perspective. We pivoted to create AI Agent Studio, placed Top 5 overall, and gained visibility from senior leadership.
The full case study is password protected
Please email me if you'd like to chat! If you are a recruiter or hiring manager the password for this project is at the top of my resume :).
