FlowSync

Inspiration

As a student and developer, I noticed that project teams spend a surprising amount of time after meetings manually reviewing notes, assigning tasks, tracking blockers, and updating project boards. Valuable information often gets buried inside long discussions, causing delays and making it difficult for teams to stay aligned.

I wanted to build a solution that could act as a bridge between conversation and execution. Instead of treating meeting notes as static documents, I wanted them to become actionable project workflows automatically.

That idea became FlowSync.

What it does

FlowSync is an AI-powered workflow automation platform that transforms meeting notes into structured project actions.

Using Gemini, FlowSync automatically extracts:

• Tasks
• Blockers
• Risks
• Milestones
• Priorities
• Recommended next actions

The platform also maintains meeting history, visualizes project priorities, and creates GitLab work items to help teams move directly from discussion to execution.

How I built it

FlowSync was built using:

• Google Cloud Run for deployment and hosting
• Gemini for intelligent meeting-note analysis
• GitLab for project tracking and issue management
• Google Cloud Agent Platform for agent configuration and GitLab MCP setup
• Next.js and TypeScript for the application frontend and backend
• Cursor as the primary development environment

The workflow is straightforward: users submit meeting notes, Gemini analyzes the content, FlowSync extracts project information, and the resulting actions can be converted into GitLab work items.

The application is deployed on Google Cloud Run, allowing it to be accessed through a live hosted URL.

Challenges I ran into

One of the biggest challenges was converting unstructured meeting discussions into useful project-management information. Meeting notes can vary significantly in style, detail, and clarity, making consistent extraction difficult.

Another challenge was integrating multiple technologies into a single workflow while keeping the user experience simple. Configuring Google Cloud services, deploying to Cloud Run, integrating GitLab workflows, and ensuring everything worked together required significant iteration and testing.

I also spent considerable time refining the interface to make the extracted information immediately useful to project managers and teams.

Accomplishments that I'm proud of

I am proud that FlowSync successfully demonstrates a complete workflow from conversation to execution.

Users can:

• Analyze meeting notes using AI
• Extract tasks, blockers, risks, and milestones
• Maintain project history
• Visualize priorities
• Create GitLab work items automatically
• Access the platform through a deployed cloud application

Most importantly, FlowSync transforms a traditionally manual process into an automated workflow that saves time and improves visibility.

What I learned

This project helped me gain hands-on experience with:

• Google Cloud Run deployment
• Gemini-powered AI workflows
• GitLab integrations
• Workflow automation design
• Full-stack application development
• Building cloud-native applications under hackathon constraints

I also learned the importance of designing AI systems that provide practical value rather than simply generating information.

What's next for FlowSync

In the future, I would like to expand FlowSync into a more comprehensive project operations platform.

Potential improvements include:

• Conversational project assistants
• Enhanced GitLab workflow automation
• Advanced analytics and forecasting
• Cross-project insights
• Additional enterprise integrations
• Smarter project memory across long-running initiatives

The long-term vision is to help teams spend less time managing work and more time delivering outcomes.

Built With

• gemini
• gitlab-api
• gitlab-mcp
• google-cloud-agent-platform
• google-cloud-run
• next.js
• node.js
• react
• tailwind-css
• typescript