Trusting and learning are two verbs that are present in the daily lives of university professors, especially when it comes to digital technologies in teaching. For these professionals to teach something to their students, they need to learn the content of what will be taught and feel confident to teach it. This research aims to understand the relationship between teacher computational self-efficacy and the process of self-regulation of university professors learning to teach through digital technologies in the pandemic context. This study developed an investigation of mixed methods carried out in two phases: the first phase was quantitative, applied in 2021, and had the participation of 409 teachers from different states of Brazil, who answered the Participant Characterization Questionnaire and the Scale on Integration of Information Technology for Teaching – EITIE (ALVARENGA, 2011); and the second phase was qualitative, developed in 2022 through semistructured interviews with 12 teachers selected based on criteria created from the first phase of the research. For data analysis in the first phase, SPSS version 28 software was used and, in the second phase, Atlas Ti version 9 was used to perform a content analysis of the data. The results of the first phase indicated that the significant variables were: age, sex, degree, activity and institution. More specifically, male, younger professors, with specialization, belonging to the Exact Sciences and Engineering and public university presented higher beliefs of computational teaching self-efficacy; in contrast, older women, doctors, humanities and public universities had lower beliefs. In the second phase of the research, it was observed that the teachers digital learning process was spontaneous and occasional, while the use of these resources in teaching was self-regulated. The use of various learning strategies was still notorious, most of them being cognitive. Another important piece of data recorded was that professors with a higher belief in computational self-efficacy were those who were more open to learning about digital technologies. In the end, teaching computational self-efficacy beliefs seem to represent a key element in the process of self-regulation of university professors learning.