- Популярные видео
- Авто
- Видео-блоги
- ДТП, аварии
- Для маленьких
- Еда, напитки
- Животные
- Закон и право
- Знаменитости
- Игры
- Искусство
- Комедии
- Красота, мода
- Кулинария, рецепты
- Люди
- Мото
- Музыка
- Мультфильмы
- Наука, технологии
- Новости
- Образование
- Политика
- Праздники
- Приколы
- Природа
- Происшествия
- Путешествия
- Развлечения
- Ржач
- Семья
- Сериалы
- Спорт
- Стиль жизни
- ТВ передачи
- Танцы
- Технологии
- Товары
- Ужасы
- Фильмы
- Шоу-бизнес
- Юмор
Voice + Vision AI Property Management | Gemini Live API, Google ADK, Cloud Run
PSLY is a multimodal AI platform for property management — tenants
report issues by voice, upload photos for AI damage assessment,
technicians get hands-free job briefings, and managers watch everything
update live on one dashboard.
Built for the Gemini Live Agent Challenge hackathon.
WHAT YOU'LL SEE:
0:00 — Problem + Solution (voice + vision AI for property ops)
0:30 — Tenant: Voice report with Gemini Live API + photo upload with
Gemini Vision
1:40 — Manager: Real-time dashboard, AI severity scores, one-click
technician assignment
2:30 — Technician: Hands-free voice briefing, status updates, live sync
to tenant + manager
3:30 — Architecture overview
GOOGLE AI & CLOUD PRODUCTS USED:
• Gemini Live API — bidirectional streaming voice with barge-in support
• Gemini Vision — automated photo damage assessment (severity, type,
confidence)
• Google ADK — 6-agent orchestration (Router → Triage, Status, Schedule,
Escalation, Analytics)
• Google Cloud Run — containerized backend deployment
TECH STACK:
• Backend: NestJS + TypeScript + Prisma + PostgreSQL (Supabase)
• Frontend: React + TypeScript + Vite + Tailwind CSS
• Real-time: Socket.IO WebSockets
• AI Model: gemini-2.5-flash (text + vision),
gemini-2.5-flash-native-audio (voice)
LINKS:
🔗 Live Demo: https://frontend-theta-dusky-58.vercel.app
💻 GitHub: https://github.com/hislordshipprof/opsly
Demo accounts (password: password123):
• Tenant: tenant@opsly.io
• Manager: sarah@opsly.io
• Technician: mike@opsly.io
Solo build | #GeminiLiveAgentChallenge
Видео Voice + Vision AI Property Management | Gemini Live API, Google ADK, Cloud Run канала Benjamin Agyekum
report issues by voice, upload photos for AI damage assessment,
technicians get hands-free job briefings, and managers watch everything
update live on one dashboard.
Built for the Gemini Live Agent Challenge hackathon.
WHAT YOU'LL SEE:
0:00 — Problem + Solution (voice + vision AI for property ops)
0:30 — Tenant: Voice report with Gemini Live API + photo upload with
Gemini Vision
1:40 — Manager: Real-time dashboard, AI severity scores, one-click
technician assignment
2:30 — Technician: Hands-free voice briefing, status updates, live sync
to tenant + manager
3:30 — Architecture overview
GOOGLE AI & CLOUD PRODUCTS USED:
• Gemini Live API — bidirectional streaming voice with barge-in support
• Gemini Vision — automated photo damage assessment (severity, type,
confidence)
• Google ADK — 6-agent orchestration (Router → Triage, Status, Schedule,
Escalation, Analytics)
• Google Cloud Run — containerized backend deployment
TECH STACK:
• Backend: NestJS + TypeScript + Prisma + PostgreSQL (Supabase)
• Frontend: React + TypeScript + Vite + Tailwind CSS
• Real-time: Socket.IO WebSockets
• AI Model: gemini-2.5-flash (text + vision),
gemini-2.5-flash-native-audio (voice)
LINKS:
🔗 Live Demo: https://frontend-theta-dusky-58.vercel.app
💻 GitHub: https://github.com/hislordshipprof/opsly
Demo accounts (password: password123):
• Tenant: tenant@opsly.io
• Manager: sarah@opsly.io
• Technician: mike@opsly.io
Solo build | #GeminiLiveAgentChallenge
Видео Voice + Vision AI Property Management | Gemini Live API, Google ADK, Cloud Run канала Benjamin Agyekum
Комментарии отсутствуют
Информация о видео
16 марта 2026 г. 11:28:05
00:04:01
Другие видео канала

