devpost icon

Inspiration

Workplace scheduling is a very complex and time-consuming task. Coordinating preferences, demand, and availability over a large employee base often takes hours of manual effort and careful planning. We wanted to alleviate some of that stress for managers and schedulers by streamlining the process using artificial intelligence with automation. That's why we dove into the DAIN track, to create an agent that works seamlessly with other DAIN agents to provide a smooth onboarding and shift-planning experience.

What it does

We noticed that DAIN lacks built-in database integration, so we set out to provide databasing as a service within the DAIN ecosystem. Our project, Onboard Scheduler, allows users to easily organize, manage, and update employee data using natural language prompts. Taking advantage of AI automation, Onboard Scheduler can:

Interpret user text prompts to add or remove employees from a custom database.
Automatically generate optimized daily schedules based on employee availability.
Provide statistical insights and visual data to help managers make informed scheduling decisions.

How we built it

We built Onboard Scheduler using a combination of powerful tools and platforms to bring seamless scheduling to life:

DAIN: for app integration, agent communication, and automation across the platform.
Supabase: as our backend database solution to store and manage employee data.
Google Gemini: to intelligently interpret and manage employee availability using natural language prompts.

Challenges we ran into

Creating dynamic UI components that utilized our ever changing database was a difficult problem we managed to overcome. Other problems we had while developing was just doing a project in tech-stack neither of us were familiar with.

Accomplishments that we're proud of

We're proud we managed to integrate so many different tools to our service, and have Butterfly utilize them effectively. A lot of the intended features we had in mind managed to come to fruition by the end of the hack, such as graphs and a full-fledge database.

What we learned

We learned to create tools for a service that could be utilized by many AI agents. Through DAIN's documentation and platform we learned of the ecosystem full of all sorts of services developed to allow autonomous assistance to take place.

What's next for Onboard Scheduler DAIN Service

Increase the database to hold more employer data, such as contact info and skills. We hope to make our data more secure either using DAIN's OAuth or moving from Superbase to host locally. We also intend on broadening the scope to "project-focus" occupations to manage workload, allocate tech, and effectively utilize worker skills.