This 15-minute conference talk from Linux Foundation's BazelCon 2020 explores how Uber Elevate's research division (ATCP) developed a reproducible Research Platform using Bazel to address the challenges of research reproducibility. Learn how this methodology helps mitigate technical debt, build incrementally on previous work, create scalable experimental designs, and produce reproducible research artifacts. Speakers Sebastien Mamessier and Romain Pennec share their contributions to the Bazel ecosystem, lessons learned during implementation, and their vision for extending Bazel as a generic build tool beyond software artifacts. The presentation includes references to research on the reproducibility crisis, introducing Maneage, and artificial intelligence's reproducibility challenges.