Case Studies of Codebeamer for Embedded AI Model Development in Automotive Software

Hi ​@NM_4198246,

Thank you for your question. 

Your post appears well documented but has not yet received any response. I am replying to raise awareness. Hopefully, another community member will be able to help.

Also, feel free to add any additional information you think might be relevant. It sometimes helps to have screenshots to better understand what you are trying to do.

Thanks,
Vivek N
Community Moderation Team

I would like to add some supplementary information and share a hypothesis.

As described in the following article, it appears possible to establish a mechanism in which the training process (experiments) of AI models and their resulting artifacts are managed using MLflow and JFrog Artifactory:
https://jfrog.com/blog/jfrog-plugin-for-mlflow/

However, I believe that this tool integration lacks certain key aspects, such as:

• How the target scope of the training process (experiments) is defined, and
• Whether the defined targets have already been trained or not.

Based on this, I speculate that it may be possible to address these gaps by integrating Codebeamer with MLflow and JFrog Artifactory, enabling:

• Management of requirements for training targets,
• Definition of training targets, and
• Tracking of the achievement status of training processes (experiments).

Furthermore, I also hypothesize that the above process and its management could be conceptually similar to safety assurance under SOTIF.

With these considerations in mind, I would greatly appreciate it if anyone could share insights or case studies regarding effective use of Codebeamer in AI model development.

​@vnamboodheri 
Thank you for your advice.

Hi ​@NM_4198246 

It is possible that you are the first to have thought of this. Do you have the ability to run an experiment and be the one with the first case study? If you do, and you need help, feel free to keep using this topic as a thread to ask questions and we can try to support you the best we are able.

Thanks,

Brian

Hello ​@BKampling_SAI ,
Thank you for your encouragement and for offering your support.

I agree that there may be an opportunity here to explore a first case study, and I am currently assessing the feasibility of conducting an internal experiment to validate this approach with my customer. If we proceed, I would greatly appreciate your guidance and may follow up with specific questions as we progress.

I will continue using this thread to share updates and seek your insights where needed.

Thanks again for your support.