Goal:

Provide high-level goal statement; providing user context and expected user outcome(s) for this Epic. 2-3 sentences... 

As a user, I'd like to be able to run a long running job (e.g. for training, eval, or batch inference). This job should run in background. I should have a way to monitor the status of the job using its ID. Different APIs should behave consistently, meaning there should be some level of coherence in how LLS jobs are managed, regardless of the type of the API. (E.g. all jobs should have artifacts; status; one can inspect the job; remove it; etc.

Acceptance Criteria:

The Acceptance Criteria provides a definition of scope and the expected outcomes - from a users point of view - defines the value proposition

• One can create a post-training or eval job.
• This job is running in background.
• User can check its status.
• User can extract its artifacts when ready.
• User can remove the completed job.
• User can cancel a running job.
• API for eval, post-training, batch-inference and any other job-based workflows behaves consistently.

Open questions:

Any additional details, questions or decisions that need to be made/addressed

• The scope of this Epic may need to be clarified / Epic may need to be split into separate stages for better targeting.
• …

Most of this work is being tracked in https://github.com/meta-llama/llama-stack/issues/1587