End-user programming methods that enable users who may not have programming or robotics experience to program robots have helped lower the barrier of entry for everyday people to use robots for their needs. However, popular end-user robot programming methods such as kinesthetic teaching can still involve various user difficulties, from acclimating to an unfamiliar programming interface to maneuvering a robot effectively without encountering joint limits or singularities. Providing personalized, just-in-time assistance can help end-users avoid introducing errors and suboptimalities into their programs due to such difficulties while enhancing their overall experience during robot programming. My PhD thesis will focus on enabling assisted end-user robot programming to support users throughout the robot programming process, from authoring to editing, by leveraging users’ natural multimodal cues. This extended abstract details my past, current, and future work towards this goal.