MERN Web App with AI

I’m kicking off a brand-new web application built on the MERN stack and need it feature-complete from day one. The platform must support secure user authentication, handle real-time data streams seamlessly, and drive full e-commerce flows (catalog, cart, checkout, payments).

On top of the core build, three AI layers are essential:
• A customer-facing chatbot that can resolve common questions and escalate edge cases.
• A recommendation engine that learns from browsing and purchase behavior to surface relevant products in real time.
• Data analytics and predictive modeling to spot sales trends, forecast demand, and feed those insights back into both the bot and the recommendation system.

Clean code, modular architecture, and well-documented APIs are non-negotiable because the project will scale quickly. React with modern hooks, Redux or Context for state, Node/Express with TypeScript, and MongoDB Atlas for managed storage are the preferred tools. Dockerized micro-services, automated tests (Jest, Cypress), and CI/CD on GitHub Actions will keep deployments smooth.

Deliverables:
• End-to-end MERN application running in a cloud dev environment
• Integrated chatbot, recommendation, and predictive modules with training scripts
• REST/GraphQL API documentation and Postman collection
• Test suite with ≥90 % coverage and deployment pipeline

Acceptance criteria:
1. New users sign up, verify email, and check out a sample product without errors.
2. Chatbot resolves at least 80 % of scripted support queries in staging tests.
3. Recommendation engine lifts click-through rate by 15 % in A/B sandbox.
4. Predictive dashboard matches historical sales within ±5 % MAE over a 30-day back-test.

With these pieces in place, the app will be production-ready and primed for future feature sprints.