Model training using build artifacts

1. A method comprising:

receiving, by a service and in response to a user token corresponding to a training request having been validated via a proxy authenticator, the training request and the user token, the training request including a training job;

retrieving, by the service, a container image;

starting, by the service, a container using the container image, wherein the container is used to perform the training job; and

determining, by the service, that the training job performed using the container is completed.

2. The method of claim 1 , further comprising:

detecting a code commit at a code repository;

sending a request for a build job to a build server; and

determining that the build job is completed.

3. The method of claim 2 , further comprising:

retrieving, using the container, a build artifact generated by the build job;

performing, by a container platform using the container, the training job using the build artifact;

generating, by the training job, at least one training artifact; and

storing the at least one training artifact on a cloud storage platform.

4. The method of claim 3 , further comprising:

unpacking, by the container platform, the build artifact into a local directory within the container;

creating, by the container platform, a virtual environment within the container;

performing, by the container platform, a set of tasks related to the training job within the virtual environment; and

deleting, by the container platform, data from the local directory after performing the set of tasks.

5. The method of claim 4 , wherein the virtual environment comprises an isolated environment executing within the container, the isolated environment having a set of packages installed that are utilized for performing the set of tasks.

6. The method of claim 2 , wherein detecting the code commit comprises:

generating a hash string value corresponding to the code commit, the code commit comprising code changes that are stored at the code repository.

7. The method of claim 6 , wherein the hash string value is based at least in part on a cryptographic hash function.

8. The method of claim 2 , wherein determining that the build job is completed comprises:

receiving a request to store a build artifact, generated by the build job, at a location on a cloud storage platform.

9. The method of claim 8 , further comprising:

storing the build artifact at a location specified by the request on the cloud storage platform, the location comprising a universal resource locator corresponding to a particular folder at the cloud storage platform.

10. A system comprising:

one or more processors; and

a memory storing instructions that, when executed by at least one processor among the one or more processors, cause the at least one processor to perform operations comprising:

receiving, by a service and in response to a user token corresponding to a training request having been validated via a proxy authenticator, the training request and the user token, the training request including a training job;

retrieving, by the service, a container image;

starting, by the service, a container using the container image, wherein the container is used to perform the training job; and

determining, by the service, that the training job performed using the container is completed.

11. The system of claim 10 , wherein the operations further comprise:

detecting a code commit at a code repository;

sending a request for a build job to a build server; and

determining that the build job is completed.

12. The system of claim 11 , wherein the operations further comprise:

retrieving, using the container, a build artifact generated by the build job;

performing, by a container platform using the container, the training job using the build artifact;

generating, by the training job, at least one training artifact; and

storing the at least one training artifact on a cloud storage platform.

13. The system of claim 12 , wherein the operations further comprise:

unpacking, by the container platform, the build artifact into a local directory within the container;

creating, by the container platform, a virtual environment within the container;

performing, by the container platform, a set of tasks related to the training job within the virtual environment; and

deleting, by the container platform, data from the local directory after performing the set of tasks.

14. The system of claim 13 , wherein the virtual environment comprises an isolated environment executing within the container, the isolated environment having a set of packages installed that are utilized for performing the set of tasks.

15. The system of claim 11 , wherein detecting the code commit comprises:

generating a hash string value corresponding to the code commit, the code commit comprising code changes that are stored at the code repository.

16. The system of claim 15 , wherein the hash string value is based at least in part on a cryptographic hash function.

17. The system of claim 11 , wherein determining that the build job is completed comprises:

receiving a request to store a build artifact, generated by the build job, at a location on a cloud storage platform.

18. A non-transitory machine-readable medium comprising instructions which, when read by a machine, causes the machine to perform operations comprising:

receiving, by a service and in response to a user token corresponding to a training request having been validated via a proxy authenticator, the training request and the user token, the training request including a training job;

retrieving, by the service, a container image;

starting, by the service, a container using the container image, wherein the container is used to perform the training job; and

determining, by the service, that the training job performed using the container is completed.

19. The non-transitory machine-readable medium of claim 18 , wherein the operations further comprise:

detecting a code commit at a code repository;

sending a request for a build job to a build server; and

determining that the build job is completed.

20. The non-transitory machine-readable medium of claim 19 , wherein the operations further comprise:

retrieving, using the container, a build artifact generated by the build job;

performing, by a container platform using the container, the training job using the build artifact;

generating, by the training job, at least one training artifact; and

storing the at least one training artifact on a cloud storage platform.