From the source
Bayer Crop Science built the Agentic Decision Engine, a multi-agent natural language interface over its Connected Decision Engine, on 
- The Connected Decision Engine deploys, orchestrates, monitors and reuses ML workflows for plant breeding, but as adoption grew, explaining results meant piecing together context from APIs, documentation, metadata stores, dashboards and logs.
- ADE uses a supervisor agent that routes each question to a specialist agent, and in its first phase it resolves about 80 percent of user questions without an engineer in the loop.
- It runs 22 specialist agents in production, each scoped to one domain with its own tools and data sources, including a Troubleshooting Agent that reads CloudWatch telemetry and CDE API status through AgentCore Gateway.
- Bayer chose AgentCore for its managed runtime that isolates each user session, IAM-scoped tool access through AgentCore Gateway, and framework-agnostic design that let the team keep its own LangGraph orchestration and choice of models.
- AWS Unified Operations provided a Technical Account Manager and a generative AI Domain Specialist Engineer who worked with the ADE team through design and production in three technical standups each week.
- A diagnosis that once took over an hour of manual log-hunting and waiting for engineer support now takes under 10 minutes, according to the post.
