# AI Training in Kubernetes

Source: https://www.testifysec.com/blog/ai-training-kubernetes/

Author: Frederick Kautz

Published: 2024-11-12

Birds-of-a-Feather session tackling challenges of running AI training workloads on Kubernetes, including resource control, scheduling, and storage.

From the event archive

This is an event archive. Product names, affiliations, and demonstrations reflect the recording, not the current release. The overview is editorial context, not a transcript or a product support commitment.

- [Current TestifySec docs →](https://www.testifysec.com/docs)
- [Trust architecture →](https://www.testifysec.com/docs/cilock/trust)

## No recording linked

This page preserves the event listing. A recording or transcript has not been published here.

## Questions to take into the discussion

- Compute and scheduling choices for training workloads.
- Storage, resource control, and workload isolation.
- Questions for teams operating AI infrastructure.

Use these as an editorial guide while watching or exploring the topic. They are not quotations or a verified transcript of the session.

## Apply the ideas to a current workflow

For an implementation, start with the [shared documentation](https://www.testifysec.com/docs). CI/lock captures the work, Pushgate checks the push, and the platform manages gates and technical control evidence. Check the [support matrix](https://www.testifysec.com/docs/concepts/support-matrix) for the release and environment you intend to use.

A signed statement can establish attribution and detect alteration under a configured trust model. It does not establish that the producer reported honestly or that a whole compliance framework is satisfied. The [trust architecture](https://www.testifysec.com/docs/cilock/trust) explains those boundaries.