Dino — AI Vegas Dining Concierge

A conversational AI dining concierge for Las Vegas. Named after Dean Martin, with Rat Pack-era charm. Dino recommends restaurants, checks availability, books tables, and adds reservations to Google Calendar with insider tips.

Live: web-production-c0565.up.railway.app

What Dino does

Tell Dino what kind of night you're planning. He recommends restaurants with strong opinions, checks availability, books a table, and adds the reservation to your Google Calendar — complete with an insider tip about what to order or where to sit.

Dino orchestrates six tools through a single-agent tool loop:

Tool	Source	Status
search_restaurants	Google Maps Places API	Real
get_restaurant_details	Google Maps Places API	Real
check_availability	Booking service	Mock
create_reservation	Booking service	Mock
add_to_calendar	Google Calendar deep links	Real
get_booking_link	URL builder	Real

The booking service uses a mock implementation with a real API contract. The agent doesn't know it's a mock. Swapping in a real provider means changing the service, not the agent.

Architecture

Chat UI (HTML/CSS/JS) → FastAPI backend → Dino Agent (Claude Sonnet) → Tools

The backend is the product. The frontend is a thin client that calls POST /chat.

Project structure

dino/
├── src/
│   ├── agent/
│   │   ├── dino.py              # Agent loop + tool orchestration
│   │   ├── personality.py       # System prompt + Dino's voice
│   │   └── tools.py             # Tool definitions
│   ├── services/
│   │   ├── places.py            # Google Maps Places API
│   │   ├── booking.py           # Mock booking (real contract)
│   │   └── calendar.py          # Google Calendar deep links
│   ├── data/                    # Curated restaurant data
│   └── api/
│       ├── main.py              # FastAPI app
│       └── models.py            # Pydantic response models
├── frontend/
│   └── index.html               # Chat UI (vanilla HTML/CSS/JS)
├── tests/
├── .claude/
│   ├── commands/                # Claude Code slash commands
│   └── skills/                  # Claude Code skill files
├── CLAUDE.md                    # Claude Code project context
├── Procfile                     # Railway start command
├── requirements.txt
└── .env.example

Setup

Prerequisites

• Python 3.12+
• Anthropic API key
• Google Maps API key with Places API enabled

Install

git clone https://github.com/rthomchick/dino.git
cd dino
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

Configure

Copy .env.example to .env and add your keys:

ANTHROPIC_API_KEY=sk-ant-...
GOOGLE_MAPS_API_KEY=AIza...

Run

uvicorn src.api.main:app --reload --port 8000

Open http://localhost:8000.

Deployment

Deployed on Railway (Hobby plan, $5/month). Railway auto-deploys from the main branch on push.

Required environment variables in Railway:

• ANTHROPIC_API_KEY
• GOOGLE_MAPS_API_KEY

Tech stack

• Backend: Python, FastAPI, httpx, Pydantic
• AI: Anthropic API (Claude Sonnet)
• Discovery: Google Maps Places API
• Calendar: Google Calendar deep links
• Frontend: Vanilla HTML, CSS, JavaScript
• Hosting: Railway

Built by

Richard Thomchick — richardthomchick.com

Built during Week 13 of an AI Product Development Program. Part of a portfolio of AI-powered tools built by a PM learning to ship AI products.