In the previous example we have seen how to define dependencies between jobs. We can also use these dependencies and transfer data from one job to the other. Data can be transfered from a producer job to the consumer job. Yoy may have multiple consumers for the same producer. The consumer jobs are the ones with the depends_on in their job definitions. So if you have a job B which depends_on A then A is the producer and B the consumer.
To produce data every job can simply put some files into /infrabox/output. The content of this directory will be made available to all consumer jobs. The consumers may access the data in /infrabox/inputs/<producer_job_name>/. I.e. job A creates file hello.txt in /infrabox/output and job B depends_on A it can access the file as /infrabox/inputs/A/hello.txt.
See the Docker Job Documentation for a detailed explanation about the options of a docker job.