AI Agent Use Cases
Views —

AI Agent Use Cases

Added today

44All 189

Prompt to Prepare a Shopping Cart from a Recipe Photo

The author reports that Grok Bot receives a recipe photo, organizes the ingredients and quantities, and creates a shopping list by excluding ingredients already available at home.

Grok Bot
Life & Hobbies

1–20 of 189

Learning & CareersWalkthrough

Send Customized Job Listings from Public Job Boards to Telegram

This is an automation example that regularly searches multiple job boards and selects only job postings that meet custom criteria, then sends them to Telegram.

n8n

How it works

  1. Download the jobs-radar workflow file and related code from the GitHub repository.
  2. In the build_workflow.py file, set custom criteria such as job titles matching your resume, keywords, excluded titles, location, language, etc.
  3. After setting up, run the command python build_workflow.py to generate a new jobs-radar.workflow.json file.
  4. Copy the workflow file to your local n8n instance (n8n-local container) and import it using the import:workflow command.
  5. The n8n workflow ID (jobsradar0000001) is disabled by default; if you use the GitHub Actions scheduler, leave it disabled and activate only one to avoid duplicate notifications.
  6. Create a Telegram bot and obtain a token from @BotFather. Register TELEGRAM_TOKEN and TELEGRAM_CHAT_ID as environment variables in n8n or GitHub Actions.
  7. When the workflow runs, it collects postings from over 100 public job boards (official JSON/RSS supported) every 4 hours.
  8. Each posting is filtered based on custom scoring criteria, and duplicate notifications are prevented for 45 days after sending.
  9. Only postings with scores above the threshold (default 48 points) are sent to Telegram. The message includes company name, location, job title, brief description, score, reason, and URL.
  10. Status and sending history are recorded in STATUS.md, seen.json, and sent.json files.

Results

The author automated the process of receiving only job postings that match their desired job titles and conditions from multiple job boards via Telegram. They explained that you can set detailed criteria such as scoring thresholds, excluded keywords, location, and language, and that duplicate notification prevention and status monitoring are supported.

Notes

The n8n workflow must be imported directly into the local container, and the Telegram bot token and chat ID must be registered as environment variables. Be careful not to activate both GitHub Actions and n8n at the same time, as this may cause duplicate notifications. The supported data formats and fields may vary by job board, so some information may be missing. Scoring criteria and keywords should be customized to match your own resume.

Source

AnanasNaPrirode (GitHub) Original post ↗

Example prompt · n8n

Goal: I want to automatically receive notifications on Telegram for only job postings that match my resume from multiple public job boards (official JSON/RSS supported).

Required Inputs and Environment:
- Ability to run n8n as a local container (n8n-local)
- Telegram bot (created via @BotFather) and chat ID
- Download jobs-radar workflow and code from GitHub repository
- Python3, Node.js, Docker installed

Implementation and Setup Steps:
1. Clone the jobs-radar repository and set custom criteria in the build_workflow.py file, such as job titles matching your resume (e.g., Solution Architect, Engineering Manager), keywords, excluded titles, location, language, scoring threshold (e.g., 48 points), etc.
2. Run python build_workflow.py to generate the jobs-radar.workflow.json file.
3. Copy the jobs-radar.workflow.json file to the n8n-local container and import the workflow using n8n import:workflow --input=/tmp/jobs-radar.json.
4. Register the Telegram bot token (TELEGRAM_TOKEN) and chat ID (TELEGRAM_CHAT_ID) as n8n environment variables.
5. The n8n workflow ID (jobsradar0000001) is disabled by default; if using the GitHub Actions scheduler, leave the n8n schedule disabled. Activate only one to prevent duplicate notifications.
6. When the workflow runs, it collects postings from over 100 public job boards (official JSON/RSS supported) every 4 hours. Each posting is filtered based on custom scoring criteria, and duplicate notifications are prevented for 45 days after sending.
7. Only postings with scores above the threshold are sent to Telegram, and the message includes company name, location, job title, brief description, score, reason, and URL.
8. Status and sending history are recorded in STATUS.md, seen.json, and sent.json files.

Deliverables:
- Only job postings matching your resume and criteria automatically arrive in your Telegram chat
- Status and history can be checked via STATUS.md, seen.json, and sent.json files

Verification Conditions:
- Check if notifications actually arrive in Telegram
- Review the status and collection results for each job board in the STATUS.md file
- Check seen.json and sent.json files for duplicate notification prevention and sending history
- Test whether custom scoring criteria, keywords, and exclusion conditions are actually applied