AI-Driven Recommendation Engine for OpenSource Contributions

Open-source software (OSS) development has become a cornerstone of modern software engineering, providing developers with invaluable opportunities for skill development, portfolio building, and collaborative learning. However, a significant barrier persists for beginners: the difficulty of identifying appropriate repositories, understanding where to begin, and navigating the complexity of large-scale open-source projects — commonly referred to as the "cold start" problem. This paper presents an AI-Driven Recommendation Engine for Open-Source Contributions, a web-based intelligent platform that leverages the GitHub REST API, Natural Language Processing (NLP) via TF-IDF vectorization, and cosine similarity to match users with the most relevant open issues based on their individual skill sets, programming language proficiencies, and domain interests. Issues are ranked using a composite scoring mechanism integrating semantic similarity, difficulty classification, and repository health metrics. Experimental evaluation with 50 users and 15,000 open issues across 500 repositories demonstrates Precision@5 of 0.78 and MRR of 0.84, with a 34% improvement over collaborative filtering baselines in cold-start scenarios. The system further incorporates a structured onboarding module and dynamic contribution tracking for continuous personalization.

INDEX TERMS : Developer onboarding, open-source contribution, recommendation engine, NLP, TF-IDF, cosine similarity, GitHub API, cold start problem, issue recommendation, collaborative filtering.

Srinivas, Vishal Kumar Prasad, M. Keerthika (2026). AI-Driven Recommendation Engine for OpenSource Contributions. International Journal of Innovative Research in Multidisciplinary Education & Technology (IJIRMET), 11(3).