Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{203926,
        author = {Neel Kote and Ganesh Bhutkar and Sanskar Sachin Kotalwar and Rishikesh Bhagwant Kothavade and Abhishek Gopalrao Kotalwar},
        title = {A Personalized AI Tool Recommendation System Using Semantic Embeddings and LLMs},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {1},
        pages = {1231-1238},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=203926},
        abstract = {The fast development of artificial intelligence tools in various domains leads to information overload for the user. Consequently, it is not easy for the individual to identify tools that will suit their needs. Currently, most directories of AI tools utilize either keyword searching or manual organization methods. Such an approach fails to understand the query posed by the user and rarely updates recommendations based on the latter's behavior. In this paper, we introduce a real-time artificial intelligence tool recommendation system that can be used to solve the above-mentioned problems. There are three major stages in our system design. First, we use a combination of sentence transformer models and vector similarity search for semantic retrieval purposes. Second, we leverage large language models hosted on-premises via the Ollama framework to boost the effectiveness of ranking. Third, we offer personalized recommendations by means of a hybrid recommendation algorithm based on content-based and collaborative filtering and additional semantic similarity and engagement signals. Our approach is tested on a dataset of more than five hundred artificial intelligence tools that the authors collected and annotated manually. We present promising results for both Quality and relevance of the tools discovered. The outcome shows that the use of vector-based search along with reasoning by a large language model and personalized adaptability will help discover AI tools better.},
        keywords = {AI Tool Discovery, Collaborative Filtering, FAISS, Hybrid Recommendation, Large Language Models, Personalized Ranking, Semantic Search, Sentence Transformers.},
        month = {June},
        }