Python code to connect to Salesforce and use Open AI for embedding

connectSF.py

import os
import openai
import numpy as np
import pandas as pd
import requests
from ast import literal_eval

def lambda_handler(event, context):
    question = event.get('question', 'Default Question')

    # Set OpenAI and Salesforce credentials
    openai.api_key = os.getenv('OPENAI_API_KEY')
    sf_username = os.getenv('SF_USERNAME')
    sf_password = os.getenv('SF_PASSWORD')
    sf_security_token = os.getenv('SF_SECURITY_TOKEN')
    sf_instance = os.getenv('SF_INSTANCE')
    sf_instance_afterlogin = os.getenv('SF_INSTANCE_AFTERLOGIN')
    sf_client_id = os.getenv('SF_CLIENT_ID')
    sf_client_secret = os.getenv('SF_CLIENT_SECRET')

    # Salesforce authentication and data extraction
    auth_url = f'https://{sf_instance}/services/oauth2/token'
    query_url = f'https://{sf_instance_afterlogin}/services/data/v57.0/query?q=SELECT+Id,Name,Company,Title,LeadSource,Email,Status+FROM+Lead'
    auth_payload = {
        'grant_type': 'password',
        'client_id': sf_client_id,
        'client_secret': sf_client_secret,
        'username': sf_username,
        'password': sf_password + sf_security_token
    }
    auth_response = requests.post(auth_url, data=auth_payload)
    access_token = auth_response.json()['access_token']
    headers = {'Authorization': f'Bearer {access_token}'}
    query_response = requests.get(query_url, headers=headers)
    leads = query_response.json().get('records', [])

    # OpenAI Embedding
    content = "\n".join([f"{lead['Name']} {lead['Email']}" for lead in leads])
    content_chunks = [content[i:i + 8000] for i in range(0, len(content), 8000)]
    content_embeddings = []
    for chunk in content_chunks:
        content_embeddings += get_embeddings_for_text(chunk)
    question_embedding = get_embeddings_for_text(question)
    similarities = [cosine_similarity(question_embedding, record_embedding) for record_embedding in content_embeddings]
    results = list(zip(leads, similarities))
    results.sort(key=lambda x: x[1][0], reverse=True)
    top_results = results[:100]
    response_data = [{'Lead': result[0]['Name'], 'Similarity': round(result[1][0], 4)} for result in top_results]

    return {
        'statusCode': 200,
        'body': {'question': question, 'results': response_data}
    }

def cosine_similarity(A, B):
    return np.dot(A, B) / (norm(A) * norm(B))

def get_embeddings_for_text(input_term):
    input_vector = openai.Embedding.create(input=input_term, model="text-embedding-ada-002")
    return input_vector['data'][0]['embedding']