Search

        contents: [query],
      });
      const values = result.embeddings[0].values;
      const queryResult = (await qdrant.search("lyrics", {
        vector: values,

    "slug": "val-vibes",
    "link": "/blog/val-vibes",
    "description": "How to build semantic search with embeddings for Val Town within Val Town itself",
    "pubDate": "Tue, 18 Jun 2024 00:00:00 GMT",
    "author": "JP Posma",

    "slug": "val-vibes",
    "link": "/blog/val-vibes",
    "description": "How to build semantic search with embeddings for Val Town within Val Town itself",
    "pubDate": "Tue, 18 Jun 2024 00:00:00 GMT",
    "author": "JP Posma",

/**
 * Calculate similarity score between two texts using OpenAI embeddings
 */
export async function calculateSimilarity(text1: string, text2: string): Promise<number> {
  const response = await openai.embeddings.create({
    model: "text-embedding-3-small",
    input: [text1, text2],

- **Frontend**: HTML, CSS (Tailwind), and TypeScript with React
- **Database**: SQLite for storing resumes and job requirements
- **NLP/ML**: OpenAI embeddings for semantic matching
## Project Structure

// Search for relevant memories based on a query
async function searchMemories(query: string): Promise<Array<{ id: number; content: string; tags: string; created_at: string }>> {
  // Simple search implementation - could be improved with embeddings or more sophisticated search
  const results = await sqlite.execute(
    `SELECT * FROM ${MEMORY_TABLE} WHERE content LIKE ? OR tags LIKE ?`,

## Future Improvements
- Implement vector embeddings for more accurate memory retrieval
- Add memory management features (edit, delete, categorize)
- Improve search with semantic similarity rather than just text matching

async function main() {
  const generateEmbeddings = await pipeline("feature-extraction");
  const embeddings = await generateEmbeddings("Hello, World!");
  console.log(embeddings);
}

    "slug": "val-vibes",
    "link": "/blog/val-vibes",
    "description": "How to build semantic search with embeddings for Val Town within Val Town itself",
    "pubDate": "Tue, 18 Jun 2024 00:00:00 GMT",
    "author": "JP Posma",

    "slug": "val-vibes",
    "link": "/blog/val-vibes",
    "description": "How to build semantic search with embeddings for Val Town within Val Town itself",
    "pubDate": "Tue, 18 Jun 2024 00:00:00 GMT",
    "author": "JP Posma",