Storing Salesforce embeddings with pgvector and OpenAI

Sequin, a company that streams data from platforms like Salesforce to databases such as Postgres, is leveraging embeddings to enhance data search capabilities. By utilizing OpenAI's API, they convert customer data into vector representations, allowing for advanced semantic search rather than traditional keyword-based search. This method enables the identification of related content, such as customer feedback and product suggestions, by generating embeddings for Salesforce records and comparing them to a query's embedding. They employ tools like pgvector in Postgres for storing and querying these embeddings and use NATS for event streaming to ensure data updates are captured. Additionally, Sequin supports writing back to Salesforce, enabling the tagging and rating of notes, thereby facilitating improved product direction and customer engagement. This integration not only consolidates customer insights in a single platform but also offers a flexible and powerful alternative to Salesforce's native AI capabilities.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.