Home > AI Solutions > Gen AI > White Papers > HR-Assist: Proof of Concept RAG-Based Matching Assistant > Overview
A world in which matching team members to their best-fit career path is seamless, efficient, and effective sounds straightforward on paper but requires extensive effort to implement. At Dell Technologies, we brought this vision to life by augmenting the Mistral 7B Instruct v2 model with a private dataset containing rotation team member preferences and rotation project requirements. We then created a RAG-based chatbot called HR-Assist to tackle the complexities of more exhaustive HR tasks. This resulted in the development of a novel system that provides optimized suggestions and streamlines team member rotation placement on an intuitive browser interface.
At its heart, the HR-Assist chatbot relies heavily on data-driven decisions and automation alongside structured support mechanisms to dramatically improve the efficiency of our rotation matching process. Our approach combines processing inputs from management platform forms filled with detailed, granular data concerning team members and project submissions and funnels that into HR-Assist. By combining a private vector database and a locally-hosted LLM, the chatbot uses the data to optimally match team members to projects in rotations. The powerful synergy has transformed our internal processes, allowing us to match team members to projects with precision and efficiency and demonstrating the practical application of our broad technological solutions in a specific, high-impact HR function.