Classification : Implement cron job to update status of jobs/batch

Problem

Both the Fine-Tuning API and the Batch API in OpenAI run asynchronously. Once a request is created, the job continues processing on OpenAI’s side, and the status only updates when the job progresses (e.g., queued → running → succeeded/failed).

Currently, to know whether a fine-tuning job or batch job has completed, the user must manually call the Retrieve API for each object. This manual polling is tedious, inefficient, and not scalable.

Proposed Solution

• Set up a cron job (or background worker) that:

• Periodically calls the Retrieve API for both fine-tuning and batch jobs.

• Updates the database with the latest job status and results.

• Ensures that when users query the database, they receive up-to-date information without needing to manually poll OpenAI endpoints.

Comienza con los puntos de entrada de Fine-Tuning API y Batch API Retrieve API descritos en el issue y, a continuación, rastrea cómo se almacenan en la base de datos el estado de los jobs y sus resultados. Define el comportamiento del worker periódico para ambos tipos de jobs, incluidas las actualizaciones de estado y los resultados recuperados. Se considera terminado cuando las consultas a la base de datos exponen información actual sin polling manual.