Telegram bot to track latest drafts for Express Entry immigration program to Canada.
Start using: https://t.me/CaImmigrationBot.
- Subscribe to get notified about the latest round
- Get information about the last round
- Get information about last 50 rounds
- Get current applicants' score distribution
- Display the data on a dashboard
graph LR;
canada("🇨🇦 EE Website");
aws("☁️ AWS Lambda");
mongo[("🌿 MongoDB")];
bot("🤖 Telegram Bot");
users("💁 Users");
dashboard("📊 Dashboard")
users-->bot<-->mongo<-->aws-->canada;
users-->dashboard-->canada;
🇨🇦 Express Entry Website. Information about all express entry rounds is published on a website and is publicly available in JSON format.
☁️ AWS Lambda. JavaScript function scheduled to run every 30 minutes - fetches data from the Express Entry Website, fetched data from MongoDB, filters only new data, inserts new data into MongoDB.
🌿 MongoDB. The main data storage in this project. Namely, it stores clean Express Entry (EE) round statistics, applicant distributions, and Telegram Users (e.g. subscribers). MongoDB instance has a trigger configured - every time new data is inserted by AWS Lambda function, MongoDB sends a signal to Telegram Bot to notify its subscribers of a new round.
🤖 Telegram Bot. A bot that has some commands available to users. Fetches data from MongoDB, formats it as a message and sends it to the users upon request. Also allows users to subscibe to new round notifications.
💁 Users. Normal users and subscribers of the Telegram Bot - can send commands or subscribe to updates.
📊 Dashboard. Provides visual access to the Express Entry data. Fetches data directly from Express Entry Website. Developed in a separate repository.
The bot is deployed to fly.io free tier. The bot is working in polling mode.
Data is stored in MongoDB which is deployed at MongoDB Atlas free tier.
New data is checked with an AWS Lambda function which is invoked every 30 minutes. Overall number and duration of invocations allow to stay in the free tier.
Dashboard is deployed at Vercel free tier.
Bot is running on a NodeJS server. Deployment is done manually via fly.io CLI tool:
flyctl deployYou can check what's happenning in the logs:
flyctl logsMongoDB Atlas instance is created manually - deployment is handled by the Atlas team.
This function is deployed manually as a zip archive that includes all contents related to aws code.
zip -r deploy.zip ./src ./node_modules ./package.json ./.env ./index.jsDashboard is deployed to Vercel automatically via connected git repository.
TODO
TODO