This inventor holds 1 USPTO granted patent. Top assignee: Salesforce, Inc.. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2026
Title: Jonathan Gellin: Inventor of Locally Stored Vector Embedding Databases
Introduction
Jonathan Gellin is an inventor based in Raleigh, North Carolina, and works at Salesforce, Inc. His patent work concerns locally stored vector databases and the use of generative artificial intelligence to produce code-related responses. He is credited with 1 patent in the information provided.
Latest Patents
Gellin’s patent, “Locally-stored vector embedding databases,” describes methods, apparatuses, and computer-program products for retrieving relevant information from vector databases stored on a local machine. The disclosed method compares a query’s vector embedding with embeddings in two databases, identifies relevant information, and combines that information with the query in a prompt. A generative AI model then uses the prompt to generate a response.
Career Highlights
At Salesforce, Gellin’s listed invention addresses how locally available context can support AI-assisted code generation. The approach brings together information from separate vector databases to inform a model’s response.
Collaborations
The provided information lists Zhujun Lan and Walter Harley as Gellin’s coworkers. It does not specify whether they collaborated with him on this patent.
Conclusion
Jonathan Gellin’s patent focuses on combining locally stored vector information with generative AI to support code generation, reflecting an innovation in context-aware AI systems.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
