ado

What ado is

ado is a free, open-source Python platform for planning computational experiment campaigns and running them at scale. Its workflow covers selecting experiments, defining parameter spaces, configuring a strategy, executing runs, and analysing results. Researchers and engineers on distributed teams can collaborate, share data, and reuse measurements through memoization in the sample store. A plugin model lets Python functions extend experiments and operators. The platform documents validated schemas, dry runs, structured results, and provenance records containing resource relationships and plugin versions. Remote execution is described for Ray clusters, with Kubernetes or OpenShift deployment and a distributed MySQL backend setup. Example applications include fine-tuning and inference performance benchmarking and predictive performance model creation. Bundled agent skills can help AI coding assistants formulate campaigns, run experiments, query data, and analyse results. The project welcomes contributions through its GitHub issue tracker.