Selecting the right alloy for a specific application is one of the key problems in physical metallurgy and can be challenging due to the vast array of options and the need to consider various factors. This paper proposes an information retrieval-based chatbot, powered by a large language model (LLM), to assist metallurgists in alloy selection. The chatbot efficiently processes text from various sources, including handbooks, manuals, research papers, and multiple PDFs, providing metallurgists with comprehensive information. To address the hallucination issue inherent in LLMs, the chatbot incorporates Langchain, a technique that enhances contextualization and semantic similarity, ensuring that responses remain firmly anchored in the provided documents. The chatbot's functionalities include alloy information search, summarization, comparison, and contributing to informed recommendations. This innovative chatbot has the potential to significantly contribute to alloy selection problems faced by metallurgists, saving time and improving decision-making.