# Hospitality 2030: A Rosewood Sand Hill Hackathon: Project Gallery - **Event:** [Hospitality 2030: A Rosewood Sand Hill Hackathon](https://cerebralvalley.ai/e/rosewood-hospitality-2030) - **When:** Sat, May 16 at 9:00 AM – 10:00 PM (PDT) - **Where:** Rosewood Sand Hill, Menlo Park, CA - **Hosts:** [Cerebral Valley](https://cerebralvalley.ai/u/cv), [Rosewood Sand Hill](https://cerebralvalley.ai/u/rosewood) - **Projects:** 61 (6 placed) - **Page:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery ## Projects ### 1. Aakash N S - Sera AI Concierge This project demonstrates future of hotel booking and arrival. "Sera" is a voice-based AI concierge for Rosewood hotels to provide a personalized welcome & stay experience for regular guests. Based on the booking & flight details, and the guest's previous stay at Rosewood hotels, Sera schedules pick-up, sets the room ambience, and recommends activities at the property. ***BOOKING PAGE EXPERIENCE*** On the booking page, guests do the following simply by talking to Sera: 1. Review details about their stay, flight, pick-up etc. 2. Change their check in (arrival) & check out (departure) dates 3. Change the type of room you're staying in (e.g. Hillside, vineyard, treehouse) 4. Request room temperature, lighting, aroma, bath water changes 5. Review recommended activities & reschedule them if needed 6. learn more about the neighborhood and place of interest 7. and much more.. ***INTERACTIVE FEATURES*** The experience of talking to Sera on the booking page is almost magical: 1. As you ask for information, the page is automatically scrolled to the relevant section. 2. When you request changes, they are highlighted on the screen to offer real-time feedback. 3. You can view a transcript of your conversation with Sera as an overlay or hide it away. ***TECHNICAL STACK*** The app is built using React & Next.js, and it uses Eleven Labs for text-to-speech and speech-to-text and Claude 4.5 Sonnet as the intelligence layer. Interactivity is achieved using tools made available to the agent. - **Team:** [Aakash N S](https://cerebralvalley.ai/u/aakashns) - **GitHub:** https://github.com/aakashns/hospitality-hack - **Demo video:** https://www.youtube.com/watch?v=wYTiGp2m35k - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=13 ### 2. Vacay-nvas Activities hyper-personalization with image feed rather than exhausting chat or text - **Team:** [Bhakti K](https://cerebralvalley.ai/u/bobo123), [Harish Krishna](https://cerebralvalley.ai/u/hakr) - **GitHub:** https://github.com/bhaktipriya/vacaynvas - **Demo video:** https://www.loom.com/share/a1dda444a7d9468cb6d41f7d50b6abd2 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=14 ### 3. daylite Maestro — The Invisible Concierge. A voice-driven operations copilot for the General Manager of a luxury hotel. Maestro plugs into any MCP server the hotel already runs — property management, housekeeping, food and beverage, spa — and orchestrates across them in seconds. It works with the existing stack; it never replaces it. Problem statement #2 — The Invisible Concierge. Hyper-personalised arrival (problem #1) is the consequence of an ambient orchestration layer that actually works: listens to every staff radio call, reads the guest's full profile, and acts proactively before the GM has to think about it. What was built today (May 16, 2026): • Anthropic Claude Opus 4.7 with streaming tool_use, a mandatory reasoning scratchpad, parallel reads and serial writes, and AbortController-driven barge-in for live interruption. • ElevenLabs Conversational AI 2.0 via Custom LLM webhook with per-scenario voice parameter shifts (calm Karp, clipped Recovery, briefing Proactive). Sub-700ms end-to-end TTFB. • ElevenLabs Music API for the demo underscore — the same sponsor whose voices speak the demo also composes its soundtrack. • Model Context Protocol connections to property management, housekeeping, F&B, and spa. Any hotel running MCP-compatible operational systems can adopt Maestro without a rip-and-replace. • Editorial-luxury dashboard (Vite 6, React 19, Tailwind 4, alabaster surface, Cormorant Garamond) plus an /operator staff-mobile surface that long-press-acknowledges into a sage-glow ack loop on the GM dashboard. • Three demo scenarios — Karp arrival (personalised hand-illustrated horse card for an 8-year-old guest), Recovery (Madera reservation API fails, Maestro falls over to Mayfield autonomously, no human re-prompt), and Proactive (PMS clock advances, Maestro acts unprompted thirty minutes before the family lands). • Technical-judge surfaces: X-Ray (Alt+X) raw MCP and tool_use stream overlay, and a "Why?" audit drawer (Shift+?) showing the reasoning chain that produced every action. All data in the demo is synthetic. Disclosed inline in the dashboard header. Built solo by Dmitrii Karataev. 17 merged pull requests, reviewed by CodeRabbit, cubic-dev-ai, and claude-design-import. - **Team:** [Dmitrii Karataev](https://cerebralvalley.ai/u/kwit) - **GitHub:** https://github.com/daylite-ai/hospitality-2030-maestro - **Demo video:** https://www.youtube.com/watch?v=6Q-uu2avf2Q - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=1 ### 4. Dare A members' club chain like Rosewood already pays for Oracle OPERA. Your front desk already enters every guest preference at check-in. You're just not using that data at the moment of arrival. Sandy Chain-Recall is the AI memory layer that unlocks it. When a member walks into ANY Rosewood property — even one they've never visited — the staff tablet lights up with a personalized prep brief composed from their cross-property history. Sandy speaks the brief into the bell captain's earpiece. The guest never sees AI. The recognition is felt. Built in 8 hours at the SF Anthropic Hackathon, May 2026. - **Team:** [James Olusoga](https://cerebralvalley.ai/u/DareDev256), [Akinwale D](https://cerebralvalley.ai/u/waleee7) - **GitHub:** https://github.com/DareDev256/chain-recall - **Demo video:** https://youtu.be/XRLruyTaI_E - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=17 ### 5. Project Atelier A personalized in-suite studio where guests converse naturally about whatever is on their mind, with specialist agents that help execute tasks on the spot and an operator dashboard that lets staff see every moment land as a claimable plan and respond in real time. - **Team:** [Momin Khan](https://cerebralvalley.ai/u/maverick_khan) - **GitHub:** https://github.com/mominkhan7/Atelier - **Demo video:** https://www.youtube.com/watch?v=-WzIlLKayxk - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=21 ### 6. Attune AI Attune turns a guest's booking into a curated genome for their ideal stay. After booking, the guest receives a link and chats or swipes through paired scenes, consuming property content while unknowingly curating their own experience. By the time they arrive, a full guest genome and activity itinerary has been generated and shared with staff. Best-in-class hospitality, delivered before they walk through the door. - **Team:** [Lance J Skarada](https://cerebralvalley.ai/u/Lskarada), [Mahi Anish Jariwala](https://cerebralvalley.ai/u/mahi), [Rishabh Ranjan](https://cerebralvalley.ai/u/rranj01) - **GitHub:** https://github.com/lskarada/Rosewood - **Demo video:** https://youtu.be/9hz8L0UnPi8 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=26 ### 7. Arrivance Arrivance is obsessed with one thing: making every guest feel like their stay was made entirely for them. Before they arrive, Arrivance already knows who they are. It combines PMS history, public social signals from LinkedIn, Instagram, and X searched via Perplexity, and pre-arrival answers to build a living guest profile — then generates a fully personalised itinerary matched to their exact preferences, travel style, family, and passions. Not a template. Not a generic recommendation. A curated stay built around this specific person, arriving today. On arrival, a warm AI concierge — powered by Anthropic Claude — greets them by voice, knows their name, knows what they love, and tailors everything in real time. Ask it to swap an experience, find something for the kids, or suggest the best Pinot Noir within twenty minutes — the itinerary updates instantly, no front desk involved. And it gets better every time. Every conversation, every preference expressed, every privilege redeemed teaches Arrivance something new about that guest. Every return stay builds on the last. The profile compounds. Over time, no human concierge — at any price point — can match what Arrivance knows about this person. That is the moat. Then comes the black card moment. Every curated experience — local restaurants, wineries, spas, activities — unlocks at a preferred rate that directly offsets what the guest paid for their stay. Activate enough privileges and a $895 nightly rate effectively becomes $340. The more they explore, the more they're delighted — and the less their stay actually cost. Every redemption also earns the host an affiliate commission, making guest delight and host revenue perfectly aligned. Powered by Anthropic Claude for guest intelligence, personalised itinerary generation, and live concierge chat. ElevenLabs for personalised welcome audio. Perplexity for real-time social and web discovery. One guest. One stay. One experience that feels completely irreplaceable. That's Arrivance. Built by Erwan Le Roy — founder of Bijou Getaways (bijougetaways.com), a luxury short-term rental portfolio of 42 listings across St. John USVI, Maui, Napa Valley, Miami, and Paris, and founder of HostProfit.ai, an AI revenue intelligence platform for STR operators. Arrivance was built and piloted directly on the Bijou Getaways portfolio — every feature was tested against real guests, real properties, and real revenue data. Erwan built Arrivance because he lived the problem firsthand: extraordinary properties, but no scalable way to deliver the personalised arrival experience every guest deserves. This is the tool he needed and couldn't find. - **Team:** [Erwan Le Roy](https://cerebralvalley.ai/u/Erwan_STR) - **GitHub:** https://github.com/bijou-web/arrivance-demo - **Demo video:** https://www.loom.com/share/d9e2bda5105f40b2943944fa751aca09 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=32 ### 8. TigerSheriff (Guest is God) An agentic AI hospitality concierge for Rosewood Sand Hill, Menlo Park and the global Rosewood portfolio. Guests get personalized, voice-first guidance booking, reservation in their native language—weather, dining, FX, card perks, trip planning, spa and dining bookings, and geofenced arrival—through a custom Python agent runtime (no third-party agent framework) with ElevenLabs voice and optional edge-tts fallback. Short-term and long-term agentic memory are used extensively end to end: every lobby concierge turn carries session context forward, while returning guests get prior preferences and trip notes recalled across visits—then written back after each briefing for the next stay. Runs fully on your laptop by default: Ollama LLM, local JSON storage, DuckDuckGo search. Set DEPLOYMENT_MODE=azure for Azure OpenAI + Cosmos DB when you need cloud scale. - **Team:** [AKRAM SHERIFF](https://cerebralvalley.ai/u/akram-AI), [P Kumar](https://cerebralvalley.ai/u/pkmr) - **GitHub:** https://github.com/akramIOT/agentic-hospitality-concierge - **Demo video:** https://www.youtube.com/watch?v=ApoIBk9Y-WU - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=33 ### 9. Rosewalk This project illustrates the basic capabilities of the hotel room configuration for the guests and as well as an engaging way to market rosewood products and services (rooms) thru reward collection and points in a virtual game that uses voice AI and Blockchain technology. - **Team:** [Umit Sami](https://cerebralvalley.ai/u/umit) - **GitHub:** https://github.com/umit-memcus/rosewalk - **Demo video:** https://youtu.be/WBE1jhM7KGs - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=2 ### 10. Sagar Golla One conversation, whole hotel is listening via Master Agent and sub-agents. https://hbdev5.github.io/hospitality2030/how-it-works.html Implemented Master Concierge, with subagents to support multiple Amenities, Golf Course, Spa, room service, you can interrupt the conversation for example what coffees you have and then pick cold brew in the middle. Agent 2 For scheduling for Golf Tee time Agent 3 for Dinner Reservations. Try the number (646) 440-8480 Call or text, sometimes guest is too busy to call, we also added SMS agent. Guest simply texts or calls the number above, "Checking in at 5" One Conversation, whole restaurant is listening to recommend Golf Tee time or setup Dinner/spa reservations. to Maximize the user experience. - **Team:** [Sagar Golla](https://cerebralvalley.ai/u/hostbuddy) - **GitHub:** https://github.com/hbdev5/hospitality2030/tree/main - **Demo video:** https://youtu.be/HrZRWlby4P4 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=10 ### 11. The Passport The Passport is a luxury hospitality CRM where the central artifact is a passport — an itinerary that begins before booking, accumulates stamps across activities, holds a conversation with a single AI concierge (James, powered by Claude Opus 4.7 with tool-use, voiced by ElevenLabs), and recombines into the right post-stay outreach. The product is the loop: a post-stay message lands, the guest taps, returns to a Passport Preview pre-filled with everything we learned last time — including observations staff captured during the stay. Built on the validated insight (confirmed by the Rosewood Sand Hill front desk this morning) that every guest currently gets the same blast text. - **Team:** [Brad D](https://cerebralvalley.ai/u/brad-d) - **GitHub:** https://github.com/potable-anarchy/sand-hill-hackathon-the-passport - **Demo video:** https://youtu.be/EUjgsawZ2Oo - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=15 ### 12. Tend Tend is a relationship intelligence tool that takes fragmented staff knowledge and turns it into a unified, personal outreach experience for every guest. - **Team:** [Carter Smith](https://cerebralvalley.ai/u/csmith), [Liz Cherson](https://cerebralvalley.ai/u/lizc) - **GitHub:** https://github.com/carterwsmith/guestbook - **Demo video:** https://youtu.be/1gmbGomhLRA - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=3 ### 13. Forecourt Luxury hotels collect years of guest data, from past stays, preferences, dietary needs, and behavioral patterns. However, they fail to act on it in the moment that matters most, the pre-arrival call. Forecourt is a voice-first AI concierge that intercepts that call, listens naturally, and synthesizes across three streams: guest history, today’s flight and schedule, and real-time property context to surface the perfect anticipatory offer before the guest walks through the door. When a returning guest mentions it is their anniversary, the system has already recalled their engagement weekend stay in the corner suite, checked dining and chef context, and prepared a personalized experience before the conversation moves on. If the guest asks for something the property cannot deliver internally, Forecourt escalates it into a Request and routes it to outside partners. The Requests view shows the service type, who requested it, when they requested it, and a shortlist of local vendors to call immediately. It also supports a handoff flow to Cvent so teams who live in Cvent can view the same request there. Every action fires in real time to a live operations dashboard. Every call ends with a printed day sheet and a briefing card for staff handoff. One call. Every request and guest tracked. Zero friction. - **Team:** [Rahul Thennarasu](https://cerebralvalley.ai/u/rahulthennarasu), [Siddharth Balaji](https://cerebralvalley.ai/u/siddharth-balaji) - **GitHub:** https://github.com/RahulThennarasu/forecourt - **Demo video:** https://www.youtube.com/watch?v=F7mNED71oL4 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=4 ### 14. Casssipeia Cassiopeia is a voice-first AI concierge for luxury hospitality — an ambient intelligence layer that anticipates guest needs before they speak. It solves the hospitality industry's biggest gap: today, guest preferences live in PMS notes nobody reads in real-time, language barriers slow every interaction, and emergencies depend on staff finding the right phone number under stress. The demo covers three real scenarios: 1. Arrival — Astra greets guests in their preferred language (Mandarin ↔ English code-switch), already knowing their room temperature, pillow preference, and rideshare ETA from profile data delivered via QR scan. 2. Recommendations — real Yelp + Google Places search filtered by guest dietary needs, with Claude-powered cultural pairing that connects the guest's heritage to nearby venues. Books the reservation and dispatches rideshare in the same turn. 3. Emergency — tone shifts to calm empathy, then orchestrates four systems in parallel: nearest pediatric ER lookup, priority rideshare dispatch, hotel staff page, and elevator unlock. Astra stays on the line until the guest is safe. Makes real calls and passes the relevant information. Built in 6 hours. Makes real API calls — not mocked. - **Team:** [Veera Sandiparthi](https://cerebralvalley.ai/u/VeeraAI) - **GitHub:** https://github.com/sandhipveera/cassiopeia-concierge - **Demo video:** https://youtu.be/3_2mg8sHoQA - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=16 ### 15. Holograph Holograph is a voice-first hospitality coordination platform that helps Rosewood deliver personalized guest experiences without compromising privacy. Guests dynamically control which preferences are visible and when, while staff across languages and properties receive seamless, real-time coordination to deliver consistent, emotionally intelligent service. By reducing operational friction and preserving guest context across the portfolio, Holograph enables intuitive hospitality that feels personal, effortless, and invisible. - **Placement:** 2nd Place - **Team:** [Laurie Sartain](https://cerebralvalley.ai/u/laurie), [Valerie Yong](https://cerebralvalley.ai/u/valerieyong) - **GitHub:** https://github.com/lsarta/holograph - **Demo video:** https://youtu.be/AKc1w2mnui8 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=5 ### 16. Team Anticipators Built an Agentic System For Hyper Personalisation and Agentic Systems - **Team:** [Sandeep Vijayarao](https://cerebralvalley.ai/u/VijayaraoSandeep) - **GitHub:** https://github.com/sandeepvijayarao09/rosewood-app/ - **Demo video:** https://youtu.be/Kz-UuslEDJQ - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=18 ### 17. curated by tia The problem: Hotel personalization today is broken. It's invasive, rigid, and doesn't understand guests at the level a great hotel should. Our idea: From the moment you arrive, the hotel should already feel like it knows you. We're building three things on the guest side. Real-time concierge — an always-on service layer that adapts to your schedule as your day unfolds. Curated itineraries — your dining, activities, and experiences pre-personalized based on everything we know about you. And powering both: persistent guest memory — continuously capturing your preferences and goals, updated through every interaction with staff. On the operations side, the admin interface serves as both a live dashboard and a dynamic guest profile. Every piece of input — a staff observation, a guest conversation, a dietary request — gets routed into structured preferences and recommendations in real time. Staff can add context after any interaction, and the profile updates instantly, so the next touchpoint is already smarter than the last. One throughline: invisible personalization that feels effortless for the guest and seamless for the team. - **Team:** [Olivia Xiaodan Meng](https://cerebralvalley.ai/u/oliviammeng), [Aastha Manocha](https://cerebralvalley.ai/u/Aashi), [Douglas Kwok](https://cerebralvalley.ai/u/douglaskwok) - **GitHub:** https://github.com/douglaskwok/rosewoodhackathon - **Demo video:** https://www.loom.com/share/93462f9e0205497e89be6d78f55b19da - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=23 ### 18. The Bridge Luxury is the feeling that someone anticipated what you needed before you asked, that the person in front of you is fully present and the hotel knows and remembers you. Staff spend 30 to 40% of every shift on manual data entry. Requests get lost at shift change. The front desk is staring at a screen. The tools get in the way of the very thing luxury hotels exist to provide. We built two things to change that. 1. The Badge Staff wear it with their uniform. When a guest speaks, the badge captures the conversation in real time, understands what is being asked, and routes it to the right place instantly. The staff member never looks away from the guest. - Guest orders food at the restaurant. Goes directly to the kitchen. - Guest asks about things to do in the city. Concierge gets a notification instantly. - Towel request goes to housekeeping. Broken shower goes to maintenance. - Staff message any teammate or department on Slack, WhatsApp, or Teams just by speaking. When a ticket comes in, the badge vibrates. 2. Guest Intelligence Layer The moment a booking lands, an AI agent activates. It searches Instagram, LinkedIn, news, and past stay history. A full guest brief is pushed to the team on Slack or Teams immediately. Five minutes before arrival, every relevant staff member gets a reminder. - Every interaction during the stay is logged back into the PMS automatically. Opera, Protel, Mews, Golden Profiles, Bonvoy. Nobody touches a keyboard. - The guest mentioned their favorite author when they booked. The book is on the nightstand when they walk in. Nobody had to ask. The experience feels more human. Because the people delivering it finally have the context to do so. Check the badge concept here: https://gamma.app/docs/The-Bridge-Rosewood-Sand-Hill-Hackathon-jeqcot4qrmzceep?mode=doc - **Placement:** Finalist - **Team:** [Vlad Senkevich](https://cerebralvalley.ai/u/vsenkevich), [Elias Pfeffer](https://cerebralvalley.ai/u/EliasPfeffer), [Kristian Dhimitri](https://cerebralvalley.ai/u/KristianDhimi17), [Benedikts Bolsakovs](https://cerebralvalley.ai/u/Bochok) - **GitHub:** https://github.com/kristerus/rosewoodHackathon - **Demo video:** https://youtu.be/9b-EUFydU5w - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=24 ### 19. Zauri Microsoft teams chatbot that collects guest activities and makes experience recommondations - **Team:** [Gus Jenkins](https://cerebralvalley.ai/u/gusdotcom) - **GitHub:** https://github.com/jenkins6g/rosewood-hospitality-hackathon - **Demo video:** https://youtu.be/ETcSlGF1zt8 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=27 ### 20. redthread.boutique Red Thread — the agentic operating system for luxury hospitality Sonia Cheng calls "A Sense of Place" the red thread that runs through every Rosewood. We built it — an agentic OS that threads guest, property, and place across the entire stay, with the guest holding the dial. Live at app.redthread.boutique. Type a reservation. Watch nine steps of real Claude work stream into the dashboard: CRM, web research on LinkedIn / X / press (gated by Hold the Thread), an AviationStack flight call, then three Haiku reasoning passes on luggage / customs / transit composing the porte cochère ETA. A full Claude Opus + Haiku Discretion pass produces the brief — bio, conversation hooks, handle-with-care, room state, welcome amenity, itinerary — each line citing where it came from. Click Agent handoff and you'll watch a real two-machine A2A conversation: Threadkeeper (Rosewood-side, on Vercel) negotiates with elias (the founder's personal AI assistant, running on a Mac Mini in his house, exposed via a Cloudflare tunnel at elias.redthread.boutique). Real Claude on each side, structured manifest extracted, dashboard rerenders with the negotiated preferences. The protocol manifest is published at /.well-known/agent.json — anyone can post. Three Rosewood problem statements, one continuous thread: - Hyper-Personalized Arrival Orchestration → real-time research + ETA + room/amenity/itinerary commits - Invisible Concierge → the Discretion Layer is a separate, auditable Claude pass; Hold the Thread (0–10) is set by the guest and gates what we even collect - Memory Engine → cross-property dossier, six-phase live thread runs Pre-arrival → Arrive → Check-in → Stay → Checkout → Post-stay - **Placement:** 1st Place - **Team:** [Benjamin Shyong](https://cerebralvalley.ai/u/BenjaminBear) - **GitHub:** https://github.com/benikigai/redthread - **Demo video:** https://youtu.be/6eoA6SxuUSg - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=28 ### 21. Rosewood Living Profile A state-of-the-art AI operating system for luxury hotels that creates continuity across every stay by turning consent-aware guest memory into coordinated staff workflows and personalized recommendations. - **Placement:** 3rd Place - **Team:** [Jerry K](https://cerebralvalley.ai/u/j3rry), [William Perocchi](https://cerebralvalley.ai/u/wpero) - **GitHub:** https://github.com/jerrykhong/rosewood-hackathon - **Demo video:** https://drive.google.com/file/d/1MyoRcYFMdMRgW0p27tU7o_Wr59d43Jaj/view?usp=sharing - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=29 ### 22. Anora - RoseField Rosefield is a multi-agent reasoning layer that prepares a luxury hotel for each guest before the guest ever asks for anything. From three explicit inputs — date of birth, place of origin, and a single sentence of intent — eight specialized agents (Conversation, Intake, Profile Hub, Property Experience Hub, Cultural Reasoning, Composition, Action Routing, Memory) collaborate to turn cultural and emotional context into concrete, routed staff actions. **The product**. One Cultural Resonance substrate, three asymmetric surfaces: - Guest Companion — a restraint-tuned conversational thread (text + voice) that runs from booking inquiry to ~2 weeks post-stay, extracting intent through dialogue rather than forms. - Operations Theater — the same agent reasoning made fully visible to staff, with every cultural inference traced to a concrete action routed to concierge, restaurant, spa, housekeeping, valet, kitchen, or garden. - Operator Portal — a simulation environment where synthetic guest cohorts walk a property's calendar to surface experience gaps before opening day. The demo is a split-screen Companion + Theater across four scenarios (pre-arrival anniversary, mid-stay weather pivot, emotional pivot, cross-cultural flash). Guest sees ~10% of the reasoning; staff sees ~100%. That asymmetry is the architectural feature. - **Team:** [Zhirui Feng](https://cerebralvalley.ai/u/Fihezro) - **GitHub:** https://github.com/Anora-Team/Rosefield - **Demo video:** https://drive.google.com/file/d/1cve7l_02uTaSO-83jBu3DFxUk9RM7ESM/view?usp=sharing - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=31 ### 23. rosai RoseAI https://rosai.vip/ What if a hotel knew you before you checked in? Today, luxury hotels send every guest the same PDF welcome packet. Same restaurants, same spa menu, same FAQ. It doesn't matter if you're there for an anniversary or a business trip — you get the same experience. We built RoseAI — an AI concierge that creates a fully personalized, mobile-first welcome page for every guest at Rosewood Sand Hill. Here's how it works. We pull from a guest's public profile — their Spotify, their booking details, their stated interests — and in one conversation, the AI generates a complete branded experience page. "Picked for You" cards matched to their exact signals. A couples spa ritual because they're celebrating an anniversary. A Ridge Vineyards tasting because they mentioned wine. Even a custom AI-composed soundtrack built from their listening DNA — ambient melodic bass because this guest lives in Space Vibes and Dubby Dub Dub. The whole page is templatized. One prompt, any guest, every stay feels like it was designed just for them. Every guest page includes a live link back to the AI concierge — so the conversation doesn't stop at check-in. This is what luxury should feel like. Not a bigger room. A room that already knows you. RoseAI. Recognition, not repetition. - **Team:** [THANAWAT TECHAUMNUAIWIT](https://cerebralvalley.ai/u/T_T), [Josh Clawson](https://cerebralvalley.ai/u/joshclawson), [Michael Gahan](https://cerebralvalley.ai/u/mgahan), [gabby liu](https://cerebralvalley.ai/u/gabbyliu) - **GitHub:** https://github.com/thiskappaisgrey/roseai-hackathon - **Demo video:** https://www.loom.com/share/b0fa4be9f377429b9926b0a76cb9a238 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=34 ### 24. GlideOS GlideOS is a persistent, property-wide memory layer for luxury hotels. It listens across the channels staff already use, voice notes, check-in, Slack threads, restaurant and spa logs, and turns the tribal knowledge held by individual employees into "house knowledge" that hands itself off across shifts, leaves, and departures. - **Team:** [Ishwari Aghav](https://cerebralvalley.ai/u/ishwari), [Ajinkya Mulay](https://cerebralvalley.ai/u/ajinkya), [Sree Lekshmi P](https://cerebralvalley.ai/u/not_a_goddess) - **GitHub:** https://github.com/ishuaghav/GlideOS - **Demo video:** https://youtu.be/cA4et-lOMIQ - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=6 ### 25. Aura: Personal Concierge An omnichannel AI concierge for Rosewood Hotels. Guests interact via email, SMS, or phone — and the system behaves as a single, omniscient entity with full context across every channel. - **Team:** [Kelvin Ma](https://cerebralvalley.ai/u/KelvinMa) - **GitHub:** https://github.com/PureSin/rosewood-aura - **Demo video:** https://youtu.be/rUJUA6gRqsU - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=7 ### 26. Origin Table Origin Table is a mobile-first staff training app that gives hospitality teams the stories, pronunciations, and knowledge they need to deliver highly personalised service — in under five minutes before a shift. Chefs and creators can **add new training material EFFORTLESSLY** through Creator Studio: a voice memo, photo, or rough notes is all it takes, and Claude turns it into a structured training card, live in the library instantly. - **Team:** [Pheobe Sun](https://cerebralvalley.ai/u/FreeBee), [Raka Kumar](https://cerebralvalley.ai/u/raka) - **GitHub:** https://github.com/freebeenotabot/origin-table - **Demo video:** https://youtu.be/vN-fWpiS628 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=19 ### 27. Pandas StayOS: Hospitality Intelligence for the Moments Guests Remember - **Team:** [Austin Ho](https://cerebralvalley.ai/u/aicho), [Anthony Ho](https://cerebralvalley.ai/u/abho) - **GitHub:** https://github.com/aicho2023/StayOS - **Demo video:** https://www.youtube.com/watch?v=a4GxUc2xmGc - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=8 ### 28. Plore live front desk dashboard of the next 4 guests to arrive, allong with enriched data about their current state (both emotional and logistical), system will auto intervene to improve state unless a human sees and overrules. Overruling logged for future improvements to the core system. - **Team:** [Hollis Nolan](https://cerebralvalley.ai/u/hollis) - **GitHub:** https://github.com/ploreai/arrivalos - **Demo video:** https://www.loom.com/share/41a87f50b42a4b078b58d7f2da4c9190 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=11 ### 29. AI Society B Drive HAP — Hospitality Agent Protocol is an open standard that lets a guest's personal AI agent (Claude, ChatGPT, Gemini) speak to a hotel's concierge agent on the guest's behalf. The protocol runs over Anthropic's MCP, adds the semantics of a stay, and treats authorization with the rigor of OAuth — scope-bounded, time-bounded, hash-chained, with zero retention on the property side. The HAP handshake is a three-phase flow: the human authorizes the scope once, the two agents negotiate the entire stay between themselves (audible, via ElevenLabs voice; logical, via real Claude API calls), and the human confirms the negotiated outcome. The plugin installs in any Claude with one click and auto-disconnects at checkout. Every event is signed and visible on a real-time operational dashboard built for the property. HEART (Human-centric Experience Agent for Rosewood Travelers) is the reference implementation, built for Hospitality 2030: A Rosewood Sand Hill Hackathon (May 16, 2026) by AISociety. Open spec on GitHub. Rosewood defines the category. Complaints, maintenance, family check-in, and identity verification escalate to humans by design. HEART amplifies staff — it doesn't replace them. The stack ARP. - **Team:** [Jose Guillermo Chavez Ayala](https://cerebralvalley.ai/u/Guillermo), [Luis Vargas](https://cerebralvalley.ai/u/luisvargasfdz) - **GitHub:** https://github.com/jgc-a/HEART - **Demo video:** https://us06web.zoom.us/clips/share/fVtGKEkMS--objxJGkqBiA - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=20 ### 30. D2 Concierge for Luxury Hotels A single spoken sentence turns a guest's preferences into a fully planned, vetted local itinerary — with meal timing, budget logic, route previews, and seamless front-desk handoff. Built exclusively for Rosewood Sand Hill. - **Team:** [Diana PL](https://cerebralvalley.ai/u/dplazar10), [Vinod Joseph](https://cerebralvalley.ai/u/vinodjoe) - **GitHub:** https://github.com/Swiftcode1/D2-Rosie - **Demo video:** https://www.youtube.com/watch?v=8cqO6PQ5NC4 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=25 ### 31. Sahith Thupakula Rosie - turns every guest's wine interactions — across every property, globally — into a living profile that makes the next stay feel like coming home. - **Team:** [Sahith Thupakula](https://cerebralvalley.ai/u/Tubaki) - **GitHub:** https://github.com/Sahiththupakula/Hospitality2030 - **Demo video:** https://www.youtube.com/shorts/djAEzcYMFlQ - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=30 ### 32. GL - PT A staff-facing intelligence dashboard for Rosewood Hotels. Écho gives front-desk and concierge teams a calm, considered read on each guest, not a marketing prompt, but a whispered brief in their ear. How do you do AI-driven personalization without making the guest feel surveilled? - **Team:** [Priya Trivedi](https://cerebralvalley.ai/u/PT012), [Gurmehar Lugani](https://cerebralvalley.ai/u/GLPT) - **GitHub:** https://github.com/phtrivedi01-pixel/echo-rosewood2 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=36 ### 33. Solo S Operations Communication - **Team:** [Sarah M](https://cerebralvalley.ai/u/sarah) - **GitHub:** https://github.com/sarahcreatesthings/hospitality2030 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=38 ### 34. Letter Heads The Rosewood Letter is a multi-agent AI pipeline that runs silently overnight and produces a single, deeply personalized document delivered physically under the guest's door at 6AM and digitally via email with a link to their private guest page. It reasons about why a guest is there: an Intent Agent produces a visit worldview that governs every decision downstream. A World Agent pulls hyperlocal conditions, a Rhythm Agent designs the emotional arc of the day, a Discovery Agent surfaces one non-obvious local find, and a Temporal Resonance Layer finds a quietly uncanny coincidence in the calendar. The letter arrives scented, typeset, and human. When the guest scans the QR code, they reach a clean, no-navigation personal page styled to their visit mood restoration, milestone, celebration, or event. It streams their ElevenLabs voice note automatically, renders an interactive crossword where every answer is a hidden recommendation, shows a suggested nearby location with live distance and travel time, and presents a one-tap reservation panel. From there, guests select itinerary items from the letter dining notes, spa arrangements, evening plans and confirm with a single button. A confirmation lands in their inbox via SendGrid. The AI disappears. The morning remains. - **Team:** [Anmol Sharma](https://cerebralvalley.ai/u/Anmols), [Siddharth Bhat](https://cerebralvalley.ai/u/Sid42) - **GitHub:** https://github.com/Anmol-tech/RoseWood-Letter - **Demo video:** https://drive.google.com/file/d/1VoiZrzLbpZx9bNY40MulWyz3jWd5WV2X/view?usp=sharing - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=39 ### 35. Aria Rosewood Service Atelier turns fragmented cross-property guest memory into staff-ready, property-aware service intelligence. The demo shows how a global VVIP profile can be interpreted for a specific Rosewood property and trip purpose, then orchestrated into coordinated actions across Front Desk, Housekeeping, F&B, Mini Bar, Room Intelligence, and Customer Relations. AI supports human-led luxury hospitality by preparing guest insights, sensitivity guardrails, suggested service language, and department-specific actions before the guest arrives. - **Team:** [Karl Qu](https://cerebralvalley.ai/u/chuyuanqu), [Sherry Guo](https://cerebralvalley.ai/u/shixuanguo) - **GitHub:** https://github.com/quchuyuan/rosewood-service-atelier - **Demo video:** https://youtu.be/Dcu-lbUr33g - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=9 ### 36. StandardOS StandardOS trains hotel teams to deliver more consistent guest experiences and better curate the experiences that happen throughout the stay. Luxury hospitality depends on people, but onboarding today is still fragmented. New employees often learn through static SOPs, inconsistent shadowing, and outdated training materials, leading to uneven service and inconsistent guest experiences. StandardOS replaces static onboarding with live conversational rehearsal. Employees practice: - guest arrivals - service recovery - multilingual hospitality - experience curation - emotional pacing - property-specific service standards The system trains teams on how to guide guests toward the right experiences naturally, whether that is spa recovery after a long flight, dining recommendations, wellness experiences, or personalized moments throughout the stay. The future of hospitality is guest experience curators, not transactional front desks. For the first time, onboarding becomes interactive, conversational, and measurable. Hotels can train, evaluate, and certify whether employees are truly ready to deliver the brand standard before stepping in front of guests. - **Team:** [Jessie Fischer](https://cerebralvalley.ai/u/goldfisch) - **GitHub:** https://github.com/jessieguestosai/standardos-hackathon - **Demo video:** https://youtu.be/6_SIMvnN9zk - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=22 ### 37. Team Sentinel: RoseView Post-Stay Relationship Continuity for Rosewood Hotels. Collect guest's memories from their stay, build a reel for them to share to social media and create a memory bank to reference in future stays - **Team:** [Arjun Krishna](https://cerebralvalley.ai/u/theoneandarjun), [Grace Esteban](https://cerebralvalley.ai/u/gracesteban) - **GitHub:** https://github.com/arjun-krishna1/roseview - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=35 ### 38. The Germans Rose is a personal assistant for Rosewood properties, designed to be experienced on screens around the property including the guests room. The app centers on a calm, pulsating orb that guests can speak to naturally, with subtle visual states for listening, thinking, and responding. The goal is to make hotel assistance feel immediate, elegant, and ambient rather than like a conventional chat interface. The agent behind the assistant has memory, that can be further enhanced by forwarding email to its inbox, from both the guests and staff. - **Team:** [Dominic Phillips](https://cerebralvalley.ai/u/dom), [Lea Marie Korsawe](https://cerebralvalley.ai/u/lmk812) - **GitHub:** https://github.com/dominicphillips/rosewood-hackathon - **Demo video:** https://drive.google.com/file/d/1uDhaskS1K4qqoKSAfhhQ_NbyhHQnXbpB/view - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=37 ### 39. GuestFlow GuestFlow turns a returning guest's reservation into a fully choreographed, human-delivered luxury arrival, all automatically, and without crossing the line into surveillance. Check out at https://guestflow.space/ - **Team:** [pranav achar](https://cerebralvalley.ai/u/prnav), [Nithin Aruswamy](https://cerebralvalley.ai/u/nithinaru), [Tarun Yadgirkar](https://cerebralvalley.ai/u/Tarun_y) - **GitHub:** https://github.com/TarunYadgirkar/guestflow - **Demo video:** https://drive.google.com/file/d/19JV14QXvGze5nEZhaVY3ZbyvdaaMXiNK/view - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=40 ### 40. Zumi An agentic intelligence layer that is aware of guest, venue and local contexts and provides recommendation throughout a guests entire stay. - **Team:** [Jordan Cooper](https://cerebralvalley.ai/u/JordanJoeCooper), [Doug McDonald](https://cerebralvalley.ai/u/DougZumi) - **GitHub:** https://github.com/dvx-zumi/zumi-hospitality-2030 - **Demo video:** https://drive.google.com/file/d/1xo2pmdplt_bgipSBAjTdMKf4ziw_3zBS/view?usp=share_link - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=41 ### 41. MITRA MITRA is a consent-based AI companion designed for Rosewood Hotels. Unlike generic hotel chatbots, MITRA is an emotionally intelligent presence that accompanies guests through their entire journey before, during, and after their stay. The demo showcases a fully interactive single-page experience built for Rosewood Sand Hill in Menlo Park, featuring three distinct guest personas: Sam (Indian tech executive, Hindi-speaking), Ana (Italian opera patron, Italian-speaking), and John (English hedge fund director). Switching personas instantly transforms every element — voice, language, room amenities, music, dining recommendations, and emotional responses — demonstrating how deeply personalized luxury hospitality can be. Key capabilities demonstrated: Emotional Intelligence: MITRA detects guest mood (fatigued, celebratory, contemplative) and adapts voice tone, room lighting, music, and recommendations in real time Voice Personalization: ElevenLabs integration delivers persona-matched voices — calm feminine Hindi for Sam, baritone Italian for Ana, bold feminine English for John — with mood-adaptive delivery (gentle whisper when tired, playful energy when celebrating) Flight Tracking with Empathy: MITRA monitors incoming flights and responds to delays with genuine empathy, proactively adjusting reservations, room settings, and staff briefings before the guest arrives Ambient Music Generation: Real-time ambient music created via ElevenLabs Sound Effects API — healing frequencies for recovery, jazz lounge for celebration, meditation tones for wellness Multi-Agent Coordination: A visual network of 6 specialized agents (MITRA, Guest, Restaurant, Spa, Concierge, Rosewood Central) coordinating behind the scenes to fulfill requests like dinner reservations in real time Global Intelligence Network: Guest preferences travel across 8 Rosewood properties worldwide, with privacy-first consent controls ensuring the guest always owns their data Curated Room Amenities: Snacks, drinks, and tea/coffee personalized to each guest with emotionally resonant reasoning ("Vahdam Masala Chai — because you always reach for chai after a long flight. It reminds you of home.") Tech Stack: Next.js App Router, TypeScript, Tailwind CSS, shadcn/ui, Framer Motion, ElevenLabs TTS and Sound Effects API, Recharts Built for: Demonstrating how agentic AI can create emotional luxury at scale — not replacing human hospitality staff, but amplifying their ability to deliver deeply personal experiences across every Rosewood property worldwide. Category tags: AI/ML, Hospitality, Voice AI, Personalization, Agentic Systems, ElevenLabs The demo showcases a fully interactive single-page experience built for Rosewood Sand Hill in Menlo Park, featuring three distinct guest personas: Sam (Indian tech executive, Hindi-speaking), Ana (Italian opera patron, Italian-speaking), and John (English hedge fund director). Switching personas instantly transforms every element — voice, language, room amenities, music, dining recommendations, and emotional responses — demonstrating how deeply personalized luxury hospitality can be. - **Team:** [Sachin Keswani](https://cerebralvalley.ai/u/shreem) - **GitHub:** https://github.com/techstar9797/mitra - **Demo video:** https://youtu.be/X942KWBpoT0 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=42 ### 42. Omakase Sense - intelligence layer for arrivals and concierge. - **Team:** [Anirudh Goyal](https://cerebralvalley.ai/u/anirudhgoyal) - **GitHub:** https://github.com/anirudh-goyal/hospitality2030 - **Demo video:** https://youtu.be/goeyQNrOOUo - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=43 ### 43. RosePulse RosePulse is an ambient intelligence layer for Rosewood staff that helps deliver “Invisible Concierge” service without making guests feel watched or interrupted. Instead of waiting for guests to ask, RosePulse surfaces subtle, consent-aware service cues from in-stay behavior, guest preferences, arrival context, staff notes, and historical memory so hotel teams can anticipate the right moment to help. For example, RosePulse can brief staff on returning guest preferences, flag likely “delight moments,” suggest proactive concierge actions, organize voice notes, and create operational tickets that translate guest signals into discreet human service. The goal is not to replace Rosewood hospitality, but to give staff a calm, intelligent pulse on what matters most for each guest in the moment. - **Team:** [Avi Rao](https://cerebralvalley.ai/u/gratitude), [Yi Zu](https://cerebralvalley.ai/u/yizucodes), [Alex Ackerman](https://cerebralvalley.ai/u/a3ackerman) - **GitHub:** https://github.com/gratitude5dee?tab=repositories - **Demo video:** https://www.loom.com/edit/2bde063b5e6c40f0ad691061da475c85 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=44 ### 44. Aria Concierge Stop building chatbots — this agent orchestrates your entire hotel stay invisibly. This isn't another voice demo. Aria is an ambient concierge agent built on a four-part architecture that makes luxury hospitality truly invisible. While most hotels deploy chatbots that wait for guests to ask questions, Aria detects needs before guests articulate them and takes action autonomously. Here's the framework. The engine orchestrates AI models and decision logic. The tooling layer connects to booking systems, room controls, and local vendor APIs to execute real actions. The modalities layer pulls signals from CCTV, IoT sensors, NFC touchpoints, and guest CRM data across the entire stay lifecycle. The interfaces push personalized recommendations to guest apps, staff platforms, email, and WhatsApp. Watch what happens when a returning VIP's flight lands two hours early. Aria detects the pattern, cross-references past dining preferences from their Tokyo stay, adjusts room temperature to their preferred 68 degrees, pre-books their favorite corner table at the property restaurant, and briefs staff with a summary of their last three visits. The guest walks into a room that remembers them without ever filling out a preference form. This solves the core tension in ultra-luxury hospitality: guests want anticipation without surveillance, personalization without repetition. - **Team:** [Joel Koh](https://cerebralvalley.ai/u/j1x) - **GitHub:** https://github.com/clodal/aria-concierge-agent.git - **Demo video:** https://youtu.be/_cQjP3_XWLQ - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=45 ### 45. AgentRose AgentRose — an AI concierge that turns guest text messages into completed resort services, with humans supervising only the edge cases. Built for Rosewood Sand Hill, a 121-room luxury resort in Menlo Park, AgentRose runs the full guest-request loop end to end: - Proposes — reaches out to a guest over text with a specific offer (a dinner reservation, a spa slot, a car, a fix) and tracks it as a live, time-limited offer. - Understands the reply — reads the guest's response, decides whether it's a yes, a no, or a question, and answers in the resort's voice. - Closes the loop — on a yes, the service is booked and confirmed automatically; on anything unclear, it hands off to staff rather than guessing. AgentRose decides, but a guardrail keeps it from ever booking on an ambiguous reply. - Learns the guest — pulls preferences, allergies, party details, and milestones out of conversations into a single Guest 360 profile, so every future interaction is more personalized. This sits on top of a four-stream service operation — Food (restaurant, bar, in-room dining), Spa, Concierge (experiences & transport), and IT — each with its own work queue, SLA tracking, cross-team handoffs, and role-based access. The net effect: guests interact entirely by text, AgentRose handles the negotiation and confirmation, and staff focus on exceptions instead of data entry. A staff-facing AI copilot also drafts suggested replies grounded in the resort's knowledge base. Technical footing: a FastAPI/Postgres backend with a Next.js/TypeScript frontend, Claude powering the agent and copilot, and SMS delivery via Sendblue. Every AI feature degrades gracefully — it runs on deterministic logic with no API key and upgrades to Claude automatically when one is present, so the product always demos and never hard-fails. - **Team:** [Faraz Fazel](https://cerebralvalley.ai/u/Faraz23), [Annie Ostojic](https://cerebralvalley.ai/u/AnnieO), [Eric Nelson](https://cerebralvalley.ai/u/stylo_eric), [Stanley Mei](https://cerebralvalley.ai/u/SCU) - **GitHub:** https://github.com/agentrose2026/agentrose2026/tree/main/backend/app/seed - **Demo video:** https://youtube.com/shorts/3pPTPjM8oOM?feature=share - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=46 ### 46. rosely Rosely shows how an AI orchestration layer can turn scattered guest data into calm, luxury-grade service without feeling like surveillance. Luxury hotels already hold rich profiles, PMS signals, and staff know-how, but they rarely fuse in time for the moment that matters: before arrival, during a quiet evening in-room, or after checkout when loyalty is won or lost. Rosely solves that gap with a live memory → planner → composer → tone guard pipeline (Claude), turning each guest’s history into staff briefs, operational actions, and invitation-style guest messages. Multi-property personas (Emma, James, Yuki) walk through arrival orchestration, an invisible concierge, and post-stay continuity, with optional ElevenLabs voice for staff and guest touchpoints. - **Team:** [Y Liu](https://cerebralvalley.ai/u/snapfinger) - **GitHub:** https://github.com/snapfinger/rosely - **Demo video:** https://youtu.be/Mwnhfa00Rm4 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=12 ### 47. knowbot Knowcross is an internal service request and guest profile managment portal. If guests want to fix broken ac etc they call the front desk and then a staff is dispatched with a request. Knowcross also manages guest profile information and preferences. Knowbot access knowcross to manage guest requests and automatically reply to reviews with guest specific preferecnes - **Team:** [Michael Yu](https://cerebralvalley.ai/u/michaelwaves) - **GitHub:** https://github.com/michaelwaves/knowbot - **Demo video:** https://youtu.be/dsUT8UWMGsI - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=58 ### 48. SuiteTalk The most valuable information for personalized service lives in the minds of staff members who interact with guests every day. SuiteTalk allows staff members to instantly share the intel they learn from guests and rewards & recognizes them when they do - reinforcing a culture of above & beyond service. The product is already in a live active pilot at Hotel Wailea, a Relais & Chateaux property on Maui. - **Placement:** Finalist - **Team:** [Josh Ellis](https://cerebralvalley.ai/u/joshellis), [Ross Ragsdale](https://cerebralvalley.ai/u/rozar) - **GitHub:** https://github.com/r0zar/suitetalk - **Demo video:** https://youtube.com/shorts/zQM_tdQZuTw - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=47 ### 49. Samarth Living Memory is an AI operating system concept for luxury hospitality that makes hotel staff superhuman without the guest ever seeing a screen. Rather than building yet another guest-facing chatbot, Living Memory inverts the model: a multi-agent system captures ephemeral staff observations during a stay, synthesizes them with flight data and Rosewood's hyper-local PlaceMaker cultural network into a living cross-property guest memory graph, then surfaces hand-delivered moments back through butlers, maître d's, and night managers. The guest never interacts with AI directly, but rather they just feel known wherever they go across Rosewood properties worldwide. Due to integration challenges during the hackathon, we were unable to complete a working MVP, but the full system architecture, agent design, data schema, and PlaceMaker inventory are documented in the repository. - **Team:** [Samarth Hiremath](https://cerebralvalley.ai/u/Samarth-Hiremath) - **GitHub:** https://github.com/Samarth-Hiremath1/Living-Memory - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=48 ### 50. Thread - HospitalityOS Thread — the autonomous multi-agent operations brain for luxury hospitality. The problem: Luxury hospitality runs on context , who the guest is, what they prefer, which property they've stayed at before, how their day is unfolding. Today that context lives in a dozen disconnected systems (PMS, CRM, restaurant POS, valet log, spa booking, concierge notebook, flight tracker) and a hundred staff brains. When things go right it's because someone in the back office stayed an extra hour. When things go wrong, three preventable failures cost real revenue: 1. Booking-funnel revenue evaporates. A guest fills the cart, opens payment, looks at the room one more time, then closes the tab. Nobody at the property knows it happened — there's no concierge capable of calling within 90 seconds to recover the booking. Hotels typically lose 60–70% of started bookings to abandonment. 2. Operational shocks aren't orchestrated. A flight delay cascades into a rebooked dinner, a reshuffled cab pickup, a stale turn-down schedule, and a confused front desk. Every department learns about it separately, by phone, after the fact. 3. Arrivals feel generic. A returning guest checks in and the desk agent reads their name off a screen. Nothing about their last stay, their pronunciation, their preferences, or the gesture the property sent them last December is at the agent's fingertips before the guest finishes saying "Hi." The solution: Thread is two surfaces over one nervous system. An operator dashboard (Reception, Live Dashboard, Operations, Investor Metrics) shows every event animated into the UI in milliseconds. A booking clone demonstrates the recovery cascade on demand. Underneath sits a multi-agent fabric — nine specialist agents (Conductor, Booker, Reception, Outreach, Memory, Brief, Housekeeping, Maintenance, Concierge) gossiping over a single event bus and reaching the outside world exclusively through MCP tools. - **Team:** [Bharat Bhavnasi](https://cerebralvalley.ai/u/bvsbharat), [Spoorthy](https://cerebralvalley.ai/u/SupriyaKorukonda) - **GitHub:** https://github.com/bvsbharat/Thread-Cummus - **Demo video:** https://youtu.be/TroXKDH0Swc - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=49 ### 51. Rosewood_DNA osewood.DNA is a web app that generates culturally grounded hospitality blueprints for Rosewood properties: interior and spatial direction, local business partnerships, signature experiences, aligned with the brand’s “A Sense of Place” philosophy. - **Team:** [Katherine Gao](https://cerebralvalley.ai/u/Katherine912) - **GitHub:** https://github.com/Siyigao991122/rosewood - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=52 ### 52. HostReel Arriving at a new suite often means guests fumbling with unfamiliar smart controls, or missing out on personalized amenities because they simply didn't know they were there. As properties elevate the guest experience with more high-end amenities, the amount of information needed to inform guests how to use the devices and appliances in their room naturally multiplies. Imagine a guest in their airport transfer. They receive a secure Magic Link. With zero app downloads or logins, it opens a customized video walkthrough of their actual suite, recorded by their concierge. But this isn't just a video. Using our custom annotation engine, the concierge has dropped interactive hotspots on the timeline. A marker on the thermostat opens a panel: 'We’ve pre-set the room to 68 degrees, just as you requested in your previous stays.' A marker on the coffee machine If a process changes, a new amenity provided, a new appliance available, it is incredibly easy to quickly update the video on demand. , ensuring the guide stays accurate within the time frame that guests expect of a reputable, high-touch brand. Under the hood, we engineered this to be lightning-fast. The platform runs on a Next.js and TypeScript foundation with a Supabase PostgreSQL database. We leverage Cloudflare R2 for zero-latency video streaming at the edge—so the guest never sees a buffering wheel on their phone—and a custom Video.js integration to precisely render the interactive timeline markers. - **Team:** [Lily Su](https://cerebralvalley.ai/u/lilyxsu) - **GitHub:** https://github.com/LilySu/HostReel - **Demo video:** https://youtu.be/Zrzwdef0AYI - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=50 ### 53. GeeGee Luxury concierge meets invisible intelligence layer. We turn one pre-arrival voice call and consented wearable data into an invisible intelligence layer for Rosewood concierge teams - translating sleep, recovery, hormonal phase, and travel fatigue into bespoke recommendations before the guest arrives. The result is “they just knew” hospitality: the right welcome, room environment, spa treatment, dinner, and itinerary pace, delivered through human staff with AI as the quiet copilot. A guest’s body profile travels with them across every Rosewood property, so the second stay always begins where the first one ended. - **Team:** [Tavishi Gupta](https://cerebralvalley.ai/u/tavishi), [Peyton Gardipee](https://cerebralvalley.ai/u/peyton) - **GitHub:** https://github.com/pwgardipee/recovery-intelligence - **Demo video:** https://www.loom.com/share/e869f91d032a4dc58104fa59fa654818 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=51 ### 54. Hotelemetry Rosewood Intelligence is a calm AI memory layer for Rosewood: a guest fingerprint that takes the 1% VIP treatment and makes it the default for every guest, without adding new hardware. - **Team:** [Carl Vincent Kho](https://cerebralvalley.ai/u/Carl_NotANerd) - **GitHub:** https://github.com/CarlKho-Minerva/rosewood-intelligence-demo - **Demo video:** https://www.loom.com/share/7bc06dd27f20467dbbd2e08e9995efbc - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=53 ### 55. PAR level Butler helps hospitality teams remember the details guests should not have to repeat. The system connects guest preferences, dietary restrictions, favorite orders, celebrations, seating preferences, VIP notes, and service history with menu, recipe, ingredient, and operational data so staff can answer questions quickly and confidently in real time. Staff can simply ask Butler questions like “What should we know about this guest tonight?” or “What can I confidently serve this guest?” The goal is to help hospitality teams deliver more personalized, consistent, and informed service across every shift and guest interaction. - **Team:** [Keith Wong](https://cerebralvalley.ai/u/parlevel), [Betty Szeto](https://cerebralvalley.ai/u/B_Szeto), [kenneth lo](https://cerebralvalley.ai/u/klovenutres) - **GitHub:** https://github.com/Jianhuax/butlerai https://github.com/kloventures/rosewood-butlerai - **Demo video:** https://www.tella.tv/video/my-video-1-4cl0 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=54 ### 56. black coffee https://drive.google.com/file/d/1BkMAZ9NhUA7mje90oTzta_ezjzPR3JMZ/view?usp=sharing -- song link We take details that a guest has given us through a PMS conversation and combine them with details about the hotel to create a custom song. The guest hears this song when they enter the room or scan a QR code and fall in love. This is a song we wrote for Mr Arora who is comign from NYC to judge pitches in silicon valley. He is a 3rd time returning guest. [Verse 1] There's a road from New York to the bay From city gray to golden day Mr. Arora, you came in true The Rosewood's glad to welcome you [Pre-Chorus] The work is bright the miles are done You brought the future one by one Founders to meet and moves to make The valley smiles for your sake [Chorus] Welcome, Mr. Arora Rosewood is yours Sixteen acres and the California shores Mountains above you gardens below Mr. Arora this is your glow [Verse 2] Wander the paths where the olive trees lean Madera at night in a honeyed scene A cocktail waiting as the sunset rolls Mr. Arora you fit these halls [Pre-Chorus] Asaya calls when you want the calm Poolside quiet and a fire-lit balm Dawn comes soft with a cappuccino line Then back to the meetings right on time [Chorus] Welcome, Mr. Arora Rosewood is yours Sixteen acres and the California shores Mountains above you gardens below Mr. Arora this is your glow [Bridge] From the work of the day to the hush of night Every corner set in the perfect light Stay as long as the hour allows The Rosewood opens and bows to you now [Final Chorus] Welcome, Mr. Arora Rosewood is yours Sixteen acres and the California shores From the work of the day to the calm of the night Mr. Arora welcome in the light - **Team:** [Davide Di Peppe](https://cerebralvalley.ai/u/davidedp), [Serena Verma](https://cerebralvalley.ai/u/SerenaVerma) - **GitHub:** https://github.com/davidedipeppe/hospitalityhackathon - **Demo video:** https://youtu.be/zlHxRXtmjqM - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=56 ### 57. Apostoli Iris is the relationship layer between Rosewood guests and the staff who serve them. An AI engine briefs each team member with discretion, ghostwrites their personal messages in their own voice for one-tap approval, and keeps relationships warm across the time between stays. The AI never speaks to the guest. It helps the humans speak for themselves. A sense of people. Before the stay. During. After. - **Team:** [Apostoli Evreniadis](https://cerebralvalley.ai/u/apostoliev) - **GitHub:** https://github.com/apostoliev/iris - **Demo video:** https://www.loom.com/share/05894c828f1f4b8ab18b88d92a506b28 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=55 ### 58. Ucha Üchá is a voice-first dispatch system for luxury hotels. Hotel staff push to talk in plain language — "extra towels to 814, Mrs. Chen Wei" — and Üchá automatically extracts room, intent, urgency, and language, routes the request to the correct department (housekeeping, maintenance, front desk, concierge), and posts it both to a live dashboard and into the team's existing Microsoft Teams channel. Two surfaces, one source of truth. Built for properties like Rosewood Sand Hill where guests currently wait while staff manually triage requests across separate Teams channels. Üchá cuts time-to-acknowledge from the industry's 4–7 minute peak baseline to under 60 seconds, with a low-confidence review queue so nothing gets dropped when audio is unclear. - **Placement:** Finalist - **Team:** [Paritosh Kulk](https://cerebralvalley.ai/u/pkul) - **GitHub:** https://github.com/paritoshk/rosewood-ucha - **Demo video:** https://www.loom.com/share/46c69e5f8e5349328f01b5c933404a91 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=57 ### 59. Benney Benney is an AI concierge that lives in your hotel room 24/7. Without using any cameras or microphones, based on non-intrusive signals, it is able to give information to room service, help you plan your trip taking into account jet lag via the Forger99 limit-cycle oscillator model. It uses a custom trained personality-aligned neural net to determine the next thing you should do in your trip by assigning your family/group keywords (non intrustively, your size is determined at booking, you fill in some interest keywords, budget is determined by room price, and staff fills in the rest as they see fit) that makes up your persona. we then use knn to find similar family/group personas and use the neural net to simulate what they would choose based on their traits. We simulate 61k+ of these queried incredibly fast to allow you to see what similar families would have chosen. In a real scenario, we would use real family data to make this continuously more accurate and give us more family personas to simulate. Additionally, we use these plans to inform staff of when guests will come back to the room, so that room service can be timed just right. Also, it is able to take live flight data and check for delays, google maps to detect traffic to the hotel on major routes, so that staff can be there to provide a warm welcome when it's needed. P.S: Benney the cat's design & name comes from Rosewood Amsterdam, where Benney is the mascot! - **Team:** [Hari Gangadharan](https://cerebralvalley.ai/u/SquaredCube) - **GitHub:** https://github.com/Hcoder10/benney - **Demo video:** https://drive.google.com/file/d/1mlGt4bdBt4Na_YBRFPOPXGNkRdKDSY5S/view?usp=sharing - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=59 ### 60. Rosewood Intelligence Rosewood Compass (prev. Rosewood Intelligence) is a premier Guest Management dashboard intended to streamline guest experiences and enrich the staff at Rosewood with all the necessary information and actionable tasks to curate a premium experience - **Team:** [Mohit Doshi](https://cerebralvalley.ai/u/mohitdoshi) - **GitHub:** https://github.com/Mohit-Doshi/Hospitality2030 - **Demo video:** https://youtu.be/61siWL-wZ1g - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=60 ### 61. Mise en Place Mise en Place is the institutional-memory layer for luxury hospitality. When a returning guest is detected at arrival, the staff member's earpiece whispers a role-aware briefing in under a second wines remembered, sensitivities held, complaints averted. The sommelier hears different facts than the GM; a do-not-mention list is honored as a dignity-preserving redaction. Built on Claude for longitudinal multi-source synthesis (PMS, POS, housekeeping logs, incident reports across years and properties) and ElevenLabs for sub-second whisper voice. The wedge: we serve STAFF, not guests directly staff augmentation is the only AI narrative ultra-luxury brands accept. Three textured returning-guest scenarios ship in the demo. Front-of-house turnover in luxury hospitality runs 73% a year; this is the OS layer that finally lets a hotel remember. - **Team:** [Prannay Hebbar](https://cerebralvalley.ai/u/pran-ker) - **GitHub:** https://github.com/Pran-Ker/mise-en-place - **Demo video:** https://github.com/Pran-Ker/mise-en-place/releases/download/v1.0/demo.mp4 - **Project:** https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery?project=61 --- Markdown version of https://cerebralvalley.ai/e/rosewood-hospitality-2030/hackathon/gallery. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt