This inventor holds 1 USPTO granted patent and 1 published patent application. Top assignee: Microsoft Technology Licensing, LLC. 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: Raghav Magazine and Automated Prompt Optimization
Introduction
Raghav Magazine is an inventor based in Bengaluru, India. His patent work addresses ways to improve prompts for generative language models by combining automated refinement with retrieval of relevant context.
Latest Patents
Magazine is associated with one patent, 1 patent: “Automated Prompt Optimization for Generative Language Models.” The invention describes a prompt optimization service that refines initial instructions through controlled mutation and evaluation using multiple generative language models. It also uses a vector database of historical examples, encoded as embeddings, to retrieve semantically similar material for new tasks.
A dynamic prompt generator combines the refined instructions with relevant examples and task context. The system is designed to work with both large language models and smaller fine-tuned models. Configurable settings let users balance optimization quality and computational cost, while single-call mutations and fast vector searches help improve efficiency.
Career Highlights
Magazine’s patent focuses on making prompt generation more adaptable and resource-efficient across different applications. Its support for compact fine-tuned models is intended to reduce resource requirements while maintaining response quality. The supplied information identifies Microsoft Technology Licensing, LLC as his company.
Collaborations
The supplied information names Saravanakumar Rajmohan and Drishti Goel as Magazine’s coworkers. Their specific roles in the patent or related work were not provided.
Conclusion
Raghav Magazine’s patent presents an approach to generative-language-model prompting that combines automated instruction refinement, contextual retrieval, and flexible model deployment.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
