Early-warning intelligence for healthcare revenue operations.
Upstream is an early-warning intelligence system for healthcare revenue operations. It detects when payers change their behavior—denying more claims, changing policies, or creating new revenue risks—usually 30-60 days before traditional monthly reporting would catch it.
For Operators: Think of Upstream as a smoke detector for your revenue cycle. It doesn't fight the fire for you, but it tells you early when something's wrong so you can act before it becomes expensive.
- 📊 DriftWatch (Denial Rate Detection) — Detects week-over-week changes in payer denial rates. Catches when a payer who normally denies 8% suddenly denies 15%.
- 💰 DenialScope (Dollar Spike Detection) — Flags sudden increases in denial dollars by payer or reason code. Identifies $50K+ revenue leaks before they compound.
- 📁 Claim Upload & Normalization — CSV upload with automatic payer name and CPT code mapping. Works with your existing data.
⚠️ Smart Alerting — Statistical thresholds flag significant changes, not noise. Email alerts with evidence and context.- 📈 Weekly Analysis — Runs automatically. You get early signals, not month-end surprises.
Primary Users:
- Revenue Cycle Directors who need early visibility into payer behavior changes
- Billing Managers tracking denial trends and payer policy shifts
- RCM Analysts investigating root causes of revenue variance
What You'll See:
- Alerts when payers change denial behavior outside normal variance
- Evidence tables showing which claims, payers, and codes are affected
- Historical context to distinguish new issues from recurring patterns
- Actionable signals, not raw data dumps
Time to Value:
- Setup: 30 minutes (upload claims, configure alerts)
- First insight: After first weekly run (~5 days)
- Routine use: 5-minute daily check, 20-minute weekly review
Read more: See OPERATOR_GUIDE.md for detailed workflows and decision frameworks.
| Layer | Technology | Why |
|---|---|---|
| Backend | Python 3.12, Django 5.x | Rapid iteration, batteries included |
| API | Django REST Framework | Industry standard, JWT auth ready |
| Database | SQLite (dev), PostgreSQL (prod) | Simple dev, scalable prod |
| Security | django-auditlog, encrypted fields | PHI compliance ready |
| Frontend | Django Templates → React (planned) | Server-first, SPA later |
# Install dependencies
pip install -r requirements.txt
# Apply database migrations
python manage.py migrate
# Load demo data (optional - creates sample practice)
python manage.py loaddata demo_data
# Create a superuser
python manage.py createsuperuser
# Run the development server
python manage.py runserverVisit http://localhost:8000 to access the application.
# Run weekly payer drift detection for all customers
python manage.py run_weekly_payer_drift# Get JWT token
curl -X POST http://localhost:8000/api/v1/auth/token/ \
-H "Content-Type: application/json" \
-d '{"username": "your-user", "password": "your-pass"}'
# View API documentation
open http://localhost:8000/api/v1/docs/upstream/
├── models.py # Customer, ClaimRecord, DriftEvent, etc.
├── views.py # Web portal views
├── api/
│ ├── serializers.py # DRF serializers
│ ├── views.py # API viewsets
│ ├── permissions.py # Multi-tenant access control
│ └── urls.py # API routes
├── services/
│ └── payer_drift.py # Core drift detection algorithm
├── management/commands/ # CLI commands for scheduled tasks
├── fixtures/ # Demo data for onboarding
└── templates/ # Django HTML templates
Multi-tenant architecture, CSV uploads, payer drift detection, API layer
Trend visualization, custom date ranges, CPT group-level drift, payer benchmarking
SSO/SAML, role-based access, webhook integrations, audit logging dashboard
This project is in active development. See CHANGELOG.md for recent updates.
Questions? Contact the team at scale@getbyteworthy.com
Proprietary — © 2026 Byteworthy. All rights reserved.