MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202611062241 A) filed by Sharda University on May 16, 2026, for Explainable Candidate-Job Matching System.
Inventors include Bal Krishan Saraswat; Harshit Rai; Ayush Shukla; and Shams Murtaza.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: Abstract Disclosed is a system for explainable candidate-job matching using experience-based retrieval-augmented processing. The system comprises a computing device that receives at least one resume associated with a candidate and at least one job description associated with an employment role, and a server communicatively coupled to the computing device. The server parses the resume to generate parsed candidate information, extracts candidate skills, determines an estimated experience level based on work-history data, and routes the parsed candidate information to one of two retrieval-augmented processing pipelines based on the estimated experience level. The server generates semantic representations of the parsed candidate information and the job description, retrieves matching context from vector storage based on semantic similarity, generates an augmented matching prompt, and processes the augmented matching prompt using a language model to generate an explainable candidate-job matching output comprising a match score and a reasoning trace. Fig. 1 Dated May 14, 2026 / Kumar Tushar Srivastava IN/PA-3973 Agent for the Applicant
Disclaimer: Curated by HT Syndication.
