Disclaimer: This is a personal custom (in-progress) development for learning purposes.
To begin my learning journey with AI Agents, I came up with the following storyline:
An AI Agent that keeps an eye in the current shift nonconformances, provide AI-based insights and recommendations, and with some reasoning, decides and take some proactive and automatic actions, like raising alerts to collaborate with supervisors and maintenance team, or trigger actions in other solutions or external systems, if any support seems to be needed.
Today, I reached the very first milestone of my story: "Keep an eye in the current shift nonconformances and provide AI-based insights and recommendations".
To achieve that, I used SAP Digital Manufacturing, SAP AI Core and SAP Generative AI Hub, SAP BTP and python to build the following orchestration logic:
Which is deployed in SAP BTP, and called from SAP Digital Manufacturing PODs via an Action Button, to display the AI-based insights and recommendations on the fly using POD Messages, as shown below:
In my experience so far, hallucinations are rare and the insights and recommendations make sense most of the time. However, moving forward, I will explore some grounding possibilities, to reduce the risk of hallucinations and improve the overall responses.
More to come, hopefully soon! Stay tuned. 🙂
Learn more about Generative AI Hub in SAP AI Core, see diagram below:
Br, Manoel Costa
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Subject | Kudos |
---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
User | Count |
---|---|
8 | |
6 | |
4 | |
3 | |
3 | |
2 | |
2 | |
2 | |
1 | |
1 |