Modern doctoral education in engineering lies at the intersection of three topical phenomena: firstly, the surge of wicked sustainability concerns and the subsequent burgeoning demand for cross-disciplinarity. Secondly, the rapidly developing new technologies and global knowledge economy provide a thriving problem-solving potential, although – thirdly – this requires proactive and innovative collaboration beyond the scope of a single discipline. Thus, doctoral education needs new practices to ensure that students are equipped with new kinds of competencies to solve unpredictable and wicked sustainability problems. In order to reach these demands, we need to favor collaboration over competition. Here we approach these issues by presenting key findings from a five-year empirical study on doctoral education in engineering. Data were collected by using a journey mapping method on recently graduated doctors in engineering at Aalto University, Finland. Students from the examined research group were compared with a control group. The data were clustered and the main factors contributing to the individual journeys were analyzed. Community, colleagues, and collaboration turned out to have the strongest positive impact on their doctoral journey (average +1.26, scale -3…+3), and they were distributed across the thesis process. Most observations were related to external academic factors, such as funding and journal decisions. Additionally, we present research group practices, such as “Rookies club” and "Synthesis groups" that strengthened students’ resilience and internal support on these factors. These practices initiated positive interdependencies among the students and supported sustainable supervision practices. Our results are applicable to a wide range of doctoral education.