Abstract Objectives Junior doctors are frequently exposed to occupational and traumatic stress, sometimes with tragic consequences. Mindfulness-based and fitness interventions are increasingly used to mitigate this, but have not been compared. We conducted a randomised, controlled pilot trial to assess the feasibility, acceptability and effectiveness of these interventions in junior doctors. Methods We randomised participants (n=21) to weekly 1-hour sessions of personalised, trauma-informed yoga (n=10), with a 4-hour workshop, and eHealth homework; or group-format fitness (n=8) in an existing wellness programme, MDOK. Burnout, traumatic stress and suicidality were measured at baseline and 8 weeks. Results Both interventions reduced burnout, and yoga increased compassion satisfaction within group on the Professional Quality of Life scale, without difference between groups on this measure. Personalised yoga significantly reduced depersonalisation (z=−1.99, p=0.05) compared with group fitness on the Maslach Burnout Inventory (MBI-HSS (MP)) and showed greater flexibility changes. Both interventions increased MBI Personal Accomplishment, with no changes in other self-report psychological or physiological metrics, including breath-counting. Participants doing one-to-one yoga rated it more highly overall (p=0.02) than group fitness, and reported it comparatively more beneficial for mental (p=0.01) and physical health (p=0.05). Face-to-face weekly sessions were 100% attended in yoga, but only 45% in fitness. Conclusion In this pilot trial, both yoga and fitness improved burnout, but trauma-informed yoga reduced depersonalisation in junior doctors more than group-format fitness. One-to-one yoga was better adhered than fitness, but was more resource intensive. Junior doctors need larger-scale comparative research of the effectiveness and implementation of individual, organisational and systemic mental health interventions. Trial registration number ANZCTR 12618001467224.