Participatory methods, which rely heavily on community-based data collectors, are growing in popularity to deliver much-needed evidence on violence and mental health in low- and middle-income countries. These settings, along with local researchers, encounter the highest burden of violence and mental ill-health, with the fewest resources to respond. Despite increased focus on wellbeing for research participants and, to a lesser degree, professional researchers in such studies, the role-specific needs of community-based researchers receive scant attention. This co-produced paper draws insights from one group's experience to identify rewards, challenges, and recommendations for supporting wellbeing and development of community-based researchers in sensitive participatory projects in low-resource settings. Twenty-one community-based researchers supporting a mixed-methods study on youth, violence and mental health in Sri Lanka submitted 63 reflexive structured journal entries across three rounds of data collection. We applied Attride-Stirling's method for thematic analysis to explore peer researchers' learning about research, violence and mental health; personal-professional boundaries; challenges in sensitive research; and experiences of support from the core team. Sri Lanka's first study capturing experiences of diverse community-based researchers aims to inform the growing number of global health and development actors relying on such talent to deliver sensitive and emotionally difficult work in resource-limited and potentially volatile settings. Viewing participatory research as an opportunity for mutual learning among both community-based and professional researchers, we identify practice gaps and opportunities to foster respectful team dynamics and create generative and safe co-production projects for all parties. Intentional choices around communication, training, human and consumable resources, project design, and navigating instable research conditions can strengthen numerous personal and professional capacities across teams. Such individual and collective growth holds potential to benefit short- and long-term quality of evidence and inform action on critical issues, including violence and mental health, facing high-burden, low-resource contexts.