v0.1.0: Initial Release - AI-Powered SaaS Metrics Forecasting

Summary

Initial release of IntentVision - AI-powered SaaS metrics forecasting platform with multi-tenant architecture, Firestore backend, and Nixtla TimeGPT integration.

Features

• Phases 0-10: Core pipeline implementation (ingestion, normalization, forecasting, alerting)
• Phase A: Stack alignment with SaaS database tables
• Phase B: Nixtla TimeGPT integration for ML forecasting
• Phase E2E: Single-metric forecast demo with API and UI
• Phase F: Cloud deployment infrastructure (Cloud Run, Artifact Registry)
• Phase 7: Cloud Firestore wiring with live tests and CI toggle

Infrastructure

• Cloud Firestore with Application Default Credentials (ADC)
• Environment-prefixed collection paths for data isolation
• Cloud Run deployment ready with Docker support
• GitHub Actions CI/CD pipeline with test, build, deploy stages
• Live Firestore test suite (opt-in via INTENTVISION_FIRESTORE_LIVE_TESTS=1)

Packages

Package	Description
@intentvision/api	Production API server
@intentvision/contracts	Shared TypeScript contracts
@intentvision/operator	Operator dashboard and auth
@intentvision/pipeline	Data pipeline and forecasting
@intentvision/web	React web application

API Endpoints

• POST /v1/demo/ingest - Ingest metric time series data
• POST /v1/demo/forecast - Run forecast with stub or stat backend
• GET /v1/demo/metric - Retrieve metric data with latest forecast
• GET /v1/demo/backends - List available forecast backends

---

See CHANGELOG.md for full details.

🤖 Generated with Claude Code