I need to stand up an MVP-level CRM tailored for real-estate agents and teams. The core of the build is robust contact and lead management with custom tags, flexible pipeline stages, and “smart follow-ups” that automatically surface anyone who has slipped through the cracks. Automated communication is equally important. The system must be able to launch email and SMS drip sequences while also generating AI-based nudges such as “It’s been 7 days since you last contacted John…”. These reminders should respect quiet hours and unsubscribe logic. On the productivity side, the calendar module has to sync seamlessly with Google Calendar and Outlook. Agents should be able to offer real-time appointment slots, receive task reminders, and see everything reflected back in their dashboards without manual double-entry. Tech preferences are React for a mobile-friendly dashboard, a Node.js (or Python) API layer, and a PostgreSQL or MongoDB data store—feel free to recommend the best fit. Zapier hooks, Gmail/Outlook mail send, Slack notifications, and room to bolt on MLS feeds must be baked into the architecture, not treated as afterthoughts. The whole thing will launch on a scalable SaaS stack (AWS or Google Cloud are fine; SiteGround works if it meets the same bar for uptime and auto-scaling). Deliverables: • Full-stack codebase in a private repo, cleanly documented • Dev, staging, and production environments deployed • Initial database schema covering contacts, deals, tasks, campaigns, and reminders • Working integrations for email, SMS, and calendar sync • Demo data plus a quick-start script so testers can log in and see value immediately Acceptance criteria will be a live demo where an agent: 1. Imports or manually creates a lead, tags it, and places it in a pipeline stage. 2. Triggers a follow-up sequence and receives both an email and SMS. 3. Gets an AI reminder after a simulated period of inactivity. 4. Books an appointment that instantly appears in their Google Calendar. Target is a lean, production-ready MVP that can evolve into a full SaaS product, so clean code and clear modularity are essential.