Crewlytics

Inspiration

We built this because we’ve all seen what silent overload looks like. One teammate slowly burning out. Another underutilized. Managers making task decisions based on gut feeling instead of real capacity data.

Most tools track tasks. None track human bandwidth.

We wanted to build something that makes workload visible before it becomes burnout.

What it does

Our platform gives teams a real-time view of capacity.

It analyzes:

Task estimates
Skill alignment
Availability blocks
Google Calendar commitments

Then it calculates who is overloaded, who has room, and who is the smartest reassignment candidate.

Instead of reacting to missed deadlines, teams can prevent imbalance early.

How we built it

We started with the database.

We designed relational models for users, tasks, skills, and availability using PostgreSQL and Prisma. Then we built a 7-day workload engine that calculates capacity versus assigned effort.

Next, we integrated Google Calendar using OAuth. We fetch events, normalize them, and factor them into each user’s available hours.

On the frontend, we used Next.js and React to create:

A manager dashboard
An employee dashboard
A calendar sync view
Intelligent task reassignment endpoints

Everything runs through API routes with role-based access control.

Challenges we ran into

The biggest challenge was data consistency.

We ran into schema mismatches between Prisma and the database, which forced us to deeply understand migrations and synchronization. We also struggled with OAuth token handling and ensuring Google events were properly stored and refreshed.

Another major challenge was making the workload algorithm fair. We had to rethink how we weight task effort and availability so the score actually reflects reality.

Most importantly, we learned how fragile distributed systems can be — one missing column or one stale token and everything breaks.

Accomplishments we’re proud of

Fully working Google Calendar sync
Built an ML engine that predicts burnout risk and recommends task reassignments so managers can prevent overload early.
Real-time capacity computation
Smart reassignment suggestions
Clean separation between manager and employee workflows
Production-ready database structure

And we built all of this within hackathon constraints.