All projects
PROJECT NOTES / AGENTIC AI

Booking MCP

An appointment workflow that speaks your language.

A full-stack appointment demo that connects natural-language patient and doctor workflows to discoverable MCP tools and persistent scheduling data.

MCPFastAPIReactPostgreSQLDocker
AGENTIC AI↗
01Patient / doctor
02React
03FastAPI agent
04MCP server
SYSTEM ARCHITECTURE
01 / THE CHALLENGE

The problem

Appointment requests involve availability, preferences, changes, and follow-ups. A single prompt cannot carry the whole workflow without persistent context and reliable tools.

02 / THE APPROACH

How it comes together

A FastAPI application communicates with a separate MCP server. Tool discovery and invocation power availability checks, booking changes, schedule summaries, and notification wrappers.

03 / UNDER THE HOOD

Follow the flow.

SIMPLIFIED ARCHITECTURE
01Patient / doctor
02React
03FastAPI agent
04MCP server
05PostgreSQL
04 / BUILD HIGHLIGHTS

Patient chat and a doctor dashboard.

Dynamic MCP tool discovery and registration.

Book, reschedule, and cancel appointment workflows.

Session memory and prompt history restored across refreshes.

THE ENGINEERING CHOICE

Why this approach?

Maintain a real client/server boundary for MCP instead of folding tools into the web backend. This keeps tool discovery explicit and the architecture reusable.

A production-style demonstration with seeded data and calendar/email integration wrappers.

Explore the project’s supporting sources.Project repository
KEEP EXPLORING

PineRAG

Vijayshree Vaibhav
YOUR WAY INTO MY WORLD

Vijayshree’s guide.

Projects. Experience. The story behind them.

Portfolio library + Gemini when available
PORTFOLIO GUIDE

Hi! I can help you explore Vijayshree’s projects, experience, and certifications. What would you like to know?

Portfolio sources only · Gemini may process your question.