https://github.com/leonleerl/lets-video
AWS-native self-hosted video streaming platform with HLS, CloudFront, Terraform, Docker, and AI-ready workflows.
https://github.com/leonleerl/lets-video
ai-agent aws docker langgraph nextjs postgresql prisma rag react shadcn-ui tailwindcss terraform typescript video-player
Last synced: 15 days ago
JSON representation
AWS-native self-hosted video streaming platform with HLS, CloudFront, Terraform, Docker, and AI-ready workflows.
- Host: GitHub
- URL: https://github.com/leonleerl/lets-video
- Owner: leonleerl
- Created: 2026-05-10T14:19:07.000Z (about 2 months ago)
- Default Branch: main
- Last Pushed: 2026-06-08T09:23:35.000Z (29 days ago)
- Last Synced: 2026-06-08T11:10:34.239Z (29 days ago)
- Topics: ai-agent, aws, docker, langgraph, nextjs, postgresql, prisma, rag, react, shadcn-ui, tailwindcss, terraform, typescript, video-player
- Language: TypeScript
- Homepage:
- Size: 236 KB
- Stars: 1
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# LetsVideo
### AI-Powered Family Memory Platform
LetsVideo is a private family video platform designed for families living across different countries and regions.
The project was inspired by a real-world problem. My family members live in China, Australia, and Japan. Sharing videos across different messaging platforms such as WeChat, LINE, WhatsApp, and Messenger is often inconvenient. Family memories become fragmented across multiple applications, making them difficult to organise, search, and revisit over time.
LetsVideo aims to solve this problem by providing a dedicated family space where loved ones can upload, share, preserve, and rediscover their memories through videos.
In the future, LetsVideo will evolve into an AI-powered family memory platform that not only stores videos but also understands, organises, and tells the stories behind those memories.
---
# Vision
Preserve family memories across borders.
LetsVideo helps families:
* Share private videos securely
* Organise memories automatically
* Search memories naturally
* Build family timelines
* Rediscover meaningful moments
* Preserve family stories for future generations
---
# Problem Statement
Modern families are increasingly distributed across different countries.
A typical family may use:
* WeChat (China)
* LINE (Japan)
* WhatsApp (Australia)
* Google Photos
* iCloud Photos
As a result:
* Videos are scattered across multiple platforms
* Important memories become difficult to find
* Family history becomes fragmented
* Valuable moments are often forgotten
LetsVideo provides a centralised and private space dedicated to family memories.
---
# Core Features
## Family Spaces
Create a private family space and invite family members.
Examples:
* Lee Family
* Tanaka Family
* Smith Family
Each family has its own private video library.
---
## Video Sharing
Family members can:
* Upload videos
* Browse videos
* Watch videos together
* Search videos
* Organise videos
---
## Memory Timeline
Automatically organise memories by:
* Year
* Month
* Event
* Family member
Helping families revisit important moments through time.
---
# AI Agent Roadmap
The long-term goal of LetsVideo is to become an AI-powered Family Memory Agent.
## Memory Understanding Agent
When a video is uploaded, the AI Agent can:
* Understand the video content
* Generate memory descriptions
* Identify important events
* Generate searchable metadata
* Create meaningful tags
Example:
Video:
Birthday Party.mp4
Generated Memory:
"Tom celebrated her 5th birthday with her family at Hong Kong Disneyland."
---
## Memory Organisation Agent
Automatically categorise memories into:
* Travel
* Birthday
* School
* Holiday
* Family Gathering
* Wedding
No manual organisation required.
---
## Memory Search Agent
Users can ask questions such as:
* When was Tom's first trip to South Korea?
* Show me all birthday videos from 2025.
* Find videos taken in Singapore.
The AI Agent searches memories and returns relevant results.
---
## Family Story Agent
Generate:
* Annual family reviews
* Family timelines
* Growth stories
* Memory summaries
Example:
"Summarise our family's 2026 memories."
---
## Memory Documentary Agent
Future versions will help generate:
* Family documentaries
* Growth journey videos
* Event highlight reels
* Narrated memory collections
---
# Technology Stack
## Frontend
* Next.js
* TypeScript
* Tailwind CSS
## Backend
* Python
* FastAPI
## AI
* OpenAI API
* LangGraph
* LangChain
## Database
* PostgreSQL
* pgvector
## Cloud
* AWS S3
* AWS RDS
* AWS ECS
---
# Project Goals
This project is being built to explore and learn:
* AI Agents
* Function Calling
* Retrieval-Augmented Generation (RAG)
* LangGraph Workflows
* Semantic Search
* Video Metadata Understanding
* Cloud-Native Architecture
* Full-Stack AI Development
---
# Current Status
🚧 Early Development
The first milestone focuses on:
* Family spaces
* Video upload
* Video playback
* Video sharing
AI-powered memory understanding and search capabilities will be introduced in future releases.