Computer science instructors typically have many responsibilities, such as creating material, delivering lectures, clarifying student questions, and grading student deliverables, while the demand for computer science education has been increasing. Handling all of these responsibilities is challenging in itself. However, it is made worse when instructors have to spend significant time grappling with hidden obstacles and the hidden work to resolve them. While many previous studies in the CS education community have focused on improving student experience, in this work, we aim to study instructor challenges, potential practices for addressing them, and their effect on the students. The thesis of this dissertation is "ClassOps can improve educational material preparation and delivery by simultaneously addressing several needs, such as computing environment, automation, and authoring interface." This research aims to make teaching computer science less challenging for instructors and more streamlined for students. The first study found evidence for a potential lack of quality, inconsistency, and failure in an existing form of educational material, online tutorials, when analyzed through the lens of executability. We also offered possible strategies and ideas for authors and toolsmiths to improve the quality of future tutorials. In the second study, we gained more insights from the perspective of educational material authors (i.e., instructors). We reported several pain points, workarounds, and remaining desires of CS instructors and discussed preliminary ideas that may help address them in the future. In the third study, we conduct a more in-depth study on one of the pain points regarding helping students with their computing environment issues. We found the root causes of common issues students experienced in their computing environments. We inspired the design of OPUNIT, a tool for environment verification that proved effective in helping students discover and fix issues and improve confidence. Finally, we studied a potential solution to another pain point regarding course material preparation and delivery. We experimented with a computational notebook design, DOCABLE, to report its effects on the experience of instructors delivering the course and the students' learning experience. To guide the design of future tools in this space, we formalize findings from instructor and student experience as a set of properties that we deem essential in ClassOps tools in this domain. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com.bibliotheek.ehb.be/en-US/products/dissertations/individuals.shtml.]