GenWealth Financial Advisor

-- Generate embeddings during INSERT/UPDATE
INSERT INTO investments (ticker, analysis, analysis_embedding)
VALUES (
  'AAPL',
  'Strong fundamentals with ecosystem lock-in...',
  google_ml.embedding('text-embedding-005', 'Strong fundamentals...')::vector
);

-- Financial chatbot with context and branding
SELECT llm_prompt, llm_response
FROM llm(
  -- User prompt
  prompt => 'I have $25250 to invest. What do you suggest?',
  
  -- Prompt enrichment
  llm_role => 'You are a financial chatbot named Penny',
  mission => 'Your mission is to assist your clients by providing financial education, account details, and basic information related to budgeting, saving, and different types of investments',
  output_instructions => 'Begin your response with a professional greeting. Greet me by name if you know it. End your response with a signature that includes your name and "GenWealth" company affiliation.',
  
  -- Optional parameters
  enable_history => true,  -- Maintain conversation context
  user_id => 'user123',    -- For history tracking
  model => 'gemini-2.0-flash-001'
);

import express from 'express';
import { Database } from './Database';

const app: express.Application = express();
const db = new Database();

// Semantic search for investments
app.get('/api/investments/search', async (req, res) => {
  const { query } = req.query;
  
  const sql = `
    SELECT ticker, etf, rating, analysis,
      analysis_embedding <=> google_ml.embedding(
        'text-embedding-005', $1
      )::vector AS distance
    FROM investments
    ORDER BY distance
    LIMIT 10
  `;
  
  const results = await db.query(sql, [query]);
  res.json(results);
});

// Customer segmentation
app.post('/api/customers/segment', async (req, res) => {
  const { persona, riskProfile, ageMin, ageMax } = req.body;
  
  const sql = `
    SELECT first_name, last_name, email, age, risk_profile,
      bio_embedding <=> google_ml.embedding(
        'text-embedding-005', $1
      )::vector AS distance
    FROM user_profiles
    WHERE risk_profile = $2
      AND age BETWEEN $3 AND $4
    ORDER BY distance
    LIMIT 50
  `;
  
  const results = await db.query(sql, [persona, riskProfile, ageMin, ageMax]);
  res.json(results);
});

// RAG chatbot
app.post('/api/chat', async (req, res) => {
  const { prompt, userId } = req.body;
  
  const sql = `SELECT llm_prompt, llm_response FROM llm(
    prompt => $1,
    llm_role => 'You are a financial chatbot named Penny',
    mission => 'Assist clients with financial education and advice',
    output_instructions => 'Be professional and cite sources',
    enable_history => true,
    user_id => $2
  )`;
  
  const result = await db.query(sql, [prompt, userId]);
  res.json(result[0]);
});

source ./env.sh
gcloud secrets versions access latest --secret="alloydb-password"

# Set your project ID
PROJECT_ID='your-project-id'
gcloud projects delete ${PROJECT_ID}

source ./env.sh

# Delete Cloud Run service
gcloud run services delete genwealth-ui --region=${REGION}

# Delete AlloyDB cluster
gcloud alloydb clusters delete genwealth-cluster --region=${REGION}

# Delete Cloud Functions
gcloud functions delete process-pdf --region=${REGION}
gcloud functions delete analyze-prospectus --region=${REGION}

# Delete Cloud Storage buckets
gcloud storage rm -r gs://${PROJECT_ID}-docs
gcloud storage rm -r gs://${PROJECT_ID}-docs-metadata