Ask why a plant-breeding workflow failed and get an answer
Bayer Crop Science put a supervisor agent in front of 22 specialist agents so breeders and analysts can ask in plain language why a workflow failed or where a result came from. In its first phase it resolves about 80% of questions without an engineer.
- Agent or tool
- Agentic Decision Engine (Amazon Bedrock AgentCore)
- Topic
- Research & Data
- Reported on
- Blog (AWS)
- Added
- 2026-10-29
Why it matters
Research is mostly gathering, cleaning and comparing. Handing that groundwork to an agent leaves the person with the part that needs judgment: deciding what the findings mean. It is one of 71 sightings filed under Research & Data on Agent Sightings (number 2 by interest), and one of 1 involving Agentic Decision Engine (Amazon Bedrock AgentCore).
How you could try it
Here is a reusable prompt, written by us in our own words, that captures the task. Paste it into Agentic Decision Engine (Amazon Bedrock AgentCore) or any similar agent and adjust the details to your situation.
My analytics workflow failed yesterday. Read its logs and status, tell me which step failed and why in plain English, and say what I should change. If you cannot tell, say what extra information you need.
The prompt breaks down into 3 steps you can check as the agent works:
- My analytics workflow failed yesterday.
- Read its logs and status, tell me which step failed and why in plain English, and say what I should change.
- If you cannot tell, say what extra information you need.
Tip: Start small: run it once on a low-stakes example, read what the agent did, and only then let it run on real work.
Source
Read the original first-hand account on Blog (AWS): https://aws.amazon.com/blogs/machine-learning/how-bayer-crop-science-built-an-a…. Our summary is a paraphrase; details are in the source.
Featured in these guides
Built an AI agent? Feature your agent on Agent Sightings, free submission or a featured spot.