This reports on a work-in-progress of the development of an in-house corpus and high-frequency word list of the corpus for a Science and Engineering university in Japan. Students read scientific articles as part of their required English courses. However, in an informal survey, while some students were positive about the prospects of reading specialized academic articles in English, others felt that it would be too challenging. In order to bridge the difficulty gap, an in-house corpus of articles recommended by the science faculty and a high-frequency word list of the corpus are being developed. Interviews and surveys will be conducted with selected members of the science department to understand the nature of articles written in English that these members would recommend for graduate students. The articles will be gathered to create a corpus of one million words, and processed for high-frequency words using AntConc (Version 4.0.2) (Anthony, 2021) a free online vocabulary profiling software. These will be compared against the new academic word list and further analysed for specialized words. The findings will help to construct an informative vocabulary list for the students in graduate school, and in the future, this could be further refined for undergraduate students.