Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for CareFlow AI

Inspiration Home‑care agencies spend hours every week writing visit reports, incident logs, medication notes, schedules, and family updates. Caregivers are overwhelmed, and agencies struggle to stay compliant. I wanted to build something that reduces this burden and lets caregivers focus on what matters most — caring for people.

What it does CareFlow AI is an AI‑powered assistant that automates the core documentation and communication tasks in home‑care:

Generates visit reports from caregiver notes

Creates incident reports and medication logs

Suggests schedule adjustments based on constraints

Produces weekly family summaries

Stores session‑level records for quick review

It turns messy notes into clean, structured, compliant documentation in seconds.

How I built it Designed a modular agent architecture (report agent, compliance agent, scheduling agent, summary agent).

Built a Streamlit interface with multiple tabs for each workflow.

Implemented a lightweight Firestore‑style in‑memory database for demo storage.

Integrated text‑processing utilities for validation, formatting, and audio‑to‑text support.

Deployed the app on Streamlit Cloud for public access.

Challenges I ran into Structuring imports correctly for Streamlit Cloud deployment

Ensuring each agent produced consistent, reliable output

Designing a UI that felt simple but still powerful

Managing state and temporary storage without a full backend

What I learned How to design multi‑agent AI workflows

How to build and deploy Streamlit apps

How to structure a clean, scalable Python project

How to turn real‑world care‑agency workflows into automated systems

What’s next Add real Firestore or Supabase backend

Add caregiver authentication

Add voice‑to‑report pipeline

Expand compliance templates

Build a full dashboard for agencies

Built With

Share this project:

Updates

Submission history