Word Search Puzzle Generation

Aim

The idea is to leverage LLMs and serverless architecture to handle the generation of word search puzzle from start to end.

Architecture

Features

• Word Search words generation using ollama
• Word Search words generation using openai
• Puzzle generation
• Puzzle for end users (HTML)
• Puzzle for end users (PDF)
• Custom Puzzle Dimension
• Words Validation
• Safe Words

Deploying the stack

1. Ensure you have AWS CLI, AWS CDK and Docker installed
2. Run cdk deploy to deploy the stack
3. Done!

Word Generation

Prerequisite: You will to the endpoint to your ollama server. Read more on ollama here

  $ python3 scripts/ollama-word-generate.py "cartoon characters" "llama2:13b" "http://localhost:11434" > /tmp/test.json
  $ cat /tmp/test.json
    {
      "title": "Solar System",
      "0": "earth",
      "1": "mars",
      "2": "jupiter",
      "3": "saturn",
      "4": "uranus",
      "5": "neptune",
      "6": "sun",
      "7": "mercury",
      "8": "moon",
      "9": "pluto"
    }

Puzzle Generation

Note: You can test the application by uploading a list of words to the deployed S3 (/words) and see the generated puzzle in your S3 (/puzzle)

$ aws s3 cp /tmp/test.json s3://<bucket-arn>/words/test.json

Useful commands

• npm run build compile typescript to js
• npm run watch watch for changes and compile
• npm run test perform the jest unit tests
• cdk deploy deploy this stack to your default AWS account/region
• cdk diff compare deployed stack with current state
• cdk synth emits the synthesized CloudFormation template