TODO.txt

- Implement Sentiment Analysis for excerpt fetching
  - Research and choose appropriate sentiment analysis library
  - Integrate sentiment analysis into excerpt fetching process
  - Display sentiment scores alongside excerpts in UI

- Enhance verification process for automatic party/election data setting
  - Implement robust error handling and validation
  - Add logging for debugging and monitoring
  - Create unit tests to ensure reliability

- Develop Common Criticisms feature
  - Integrate with Perplexity API for generating criticisms
  - Design and implement UI for displaying criticisms
  - Add user feedback mechanism for criticism relevance

- Refactor Election Component
  - Extract state management to a separate store (e.g., using Redux or Context API)
  - Implement automatic updates for topics tab when new programs are added
  - Optimize performance with memoization techniques

- Improve build and deployment process
  - Move run.sh and deploy.sh commands into package.json scripts
  - Update documentation to reflect new build/deploy process
  - Consider implementing a CI/CD pipeline for automated deployments

- Optimize API calls for topics and parties
  - Implement parallel async calls using Promise.all()
  - Add error handling and retries for failed requests
  - Cache results to reduce unnecessary API calls

- Enhance UI/UX with ShadCN and TailwindCSS
  - Refactor components to use ShadCN where applicable
  - Optimize TailwindCSS usage for better performance
  - Ensure responsive design across all device sizes

- Set up Biome for consistent code formatting
  - Configure Biome for TypeScript and React
  - Integrate Biome with VS Code or preferred IDE
  - Add pre-commit hook to ensure code is formatted before committing

- Stream answers from Groq
- Find an elegant solution to Token Limits and CutOff of Program Content (check SambaNova or split content into multiple requests)
  - Probably best to use Gemini-Flash or Gpt-4o mini
- Make sure GroqRetry actually works