The latest articles on DEV Community by Barmaley26 (@barmaley26). https://dev.to/barmaley26 https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4078814%2F2631025e-0643-447f-9249-30ccf934b338.png https://dev.to/barmaley26 en Barmaley26 Sat, 15 Aug 2026 10:44:04 +0000 https://dev.to/barmaley26/i-built-an-ai-agent-that-answers-my-villa-booking-emails-gemini-google-adk-4p98 https://dev.to/barmaley26/i-built-an-ai-agent-that-answers-my-villa-booking-emails-gemini-google-adk-4p98 <p>I host villas in Phuket. So I built an AI agent that answers my booking emails — with Gemini and Google ADK<br> published: true</p> <h2> tags: googlecloud, ai, agents, gemini </h2> <p><em>This article was created for the purposes of entering the **All Things Agentic Hackathon</em><em>.</em></p> <h2> The chore </h2> <p>I'm a vacation-rental host in Phuket. Every week the same chain eats my hours:</p> <p>read a guest email → decode vague dates ("around Christmas", "del 29 de diciembre al 5 de enero") → check the calendar → do seasonal price math with discounts → write a warm reply → remember to place a hold → follow up.</p> <p>Every step is trivial. The chain is not. And one missed email is a lost $3,000+ booking.</p> <p>That's exactly the "messy, multi-step chore" the hackathon's Taskmaster track describes. So I built the agent I needed myself.</p> <h2> What HostPilot does </h2> <p>A raw guest inquiry goes in — email or chat text, <strong>any language</strong> — and the agent handles it end-to-end:</p> <ol> <li>Extracts dates (including cross-year stays like Dec 29 → Jan 5), party size, guest name.</li> <li>Checks real availability against the bookings store.</li> <li>Builds a quote: high/low season rates, 10% long-stay discount, cleaning fee, min-stay and capacity validation.</li> <li>If the guest clearly wants to book — <strong>places a 48-hour hold in Firestore</strong>. A real state change, not just text.</li> <li>Replies in the guest's own language (English, Spanish, Chinese in the demo) with a full price breakdown.</li> </ol> <p>My favorite moment: when requested dates conflict with an existing booking, the agent doesn't just say "unavailable" — it inspects the calendar, finds the nearest free windows on both sides, verifies and prices each one, and offers the guest two concrete alternatives.</p> <p><strong>Live demo:</strong> <a href="https://hostpilot-406193234542.asia-southeast1.run.app" rel="noopener noreferrer">https://hostpilot-406193234542.asia-southeast1.run.app</a><br> <strong>Code:</strong> <a href="https://github.com/Barmaley26/hostpilot" rel="noopener noreferrer">https://github.com/Barmaley26/hostpilot</a><br> <strong>Video:</strong> <a href="https://youtu.be/LGkHqfLt3mY" rel="noopener noreferrer">https://youtu.be/LGkHqfLt3mY</a></p> <h2> The architecture </h2> <ul> <li> <strong>Google ADK 2.6</strong> defines the agent: one <code>Agent</code>, five plain-Python tools (<code>get_property_info</code>, <code>check_availability</code>, <code>quote_price</code>, <code>create_booking</code>, <code>list_bookings</code>).</li> <li> <strong>Gemini 3.5 Flash</strong> (via Vertex AI) does the reasoning, multilingual parsing and reply generation through function calling.</li> <li> <strong>Cloud Run</strong> hosts the whole app (FastAPI + a one-screen "host inbox" UI), deployed straight from source.</li> <li> <strong>Firestore</strong> persists bookings across instances and restarts.</li> </ul> <p>guest text → FastAPI → ADK Agent (gemini-3.5-flash)<br> ├─ get_property_info<br> ├─ check_availability ──► Firestore<br> ├─ quote_price<br> └─ create_booking (48h hold) ──► Firestore<br> ← reply in guest's language + full action log</p> <h2> Three things that bit me </h2> <p><strong>1. The model has no clock.</strong> "December 20" was parsed as 2024. One line fixed year resolution, including cross-year stays:</p> <div class="highlight js-code-highlight"> <pre class="highlight plaintext"><code> python instruction = f"Today's date is {date.today().isoformat()}. ..." 2. Free tiers are not for agents. One agent turn = 5–6 generations (planning + tool calls + final reply). The Gemini API free tier (20 requests/day) died mid-testing. Switching to Vertex AI with ADC on Cloud Run solved it properly: no API keys in the container at all, billing through the project. 3. Autonomy needs guardrails in code, not in prompts. Every tool returns explicit ok/reason dicts; create_booking re-validates availability server-side. The model physically cannot double-book, no matter what it decides. The lesson Agents become reliable when the LLM plans and the tools decide. Everything that must be true — availability, price, capacity — is enforced in deterministic, testable Python. Everything that must be human — tone, language, judgment about intent — is left to Gemini. The result: a guest writing in Chinese gets a correct, warm, fully-priced booking hold in Chinese — while the host sleeps. Built solo with Gemini 3.5 Flash, Google ADK, Cloud Run and Firestore for the All Things Agentic Hackathon. </code></pre> </div> agents ai automation gemini