Trane reduces HVAC diagnostic time from 20 minutes to 20 seconds with Bedrock AgentCore
Trane Technologies built an AI‑powered conversational agent on Amazon Bedrock AgentCore and the Strands framework. The engineering team created the solution in 3–4 weeks and cut a diagnostic workflow that previously required 20 minutes of navigating multiple dashboards to a 20‑second natural‑language interaction, a 60x improvement according to internal benchmarking.
Key points
- Trane cut a 20‑minute diagnostic workflow to a 20‑second natural‑language query, a 60x speedup.
- The agentic solution was built in 3–4 weeks using Amazon Bedrock AgentCore and the Strands framework.
- AgentCore offers isolated microVM sessions, 90‑day memory expiry, and CloudWatch observability for tool‑chain tracing.
The architecture separates agent logic from tool execution, uses a centralized AgentCore Gateway for real‑time telemetry, and tailors responses for different roles such as field technicians, account managers and building owners. AgentCore provides isolated microVM sessions, built‑in session memory with a 90‑day expiry, and observability through CloudWatch, allowing the team to trace tool failures quickly. The system runs Anthropic Claude models on Bedrock to synthesize data and return answers via server‑sent events.
Trane, a global climate innovator with over $21 billion in annual revenue and operations in more than 100 countries, says the faster insights enable more proactive, data‑driven facility optimization and reduce analysis steps across its millions of connected HVAC assets.
The story so far
4 episodes →- Trane reduces HVAC diagnostic time from 20 minutes to 20 seconds with Bedrock AgentCorethis story
How Trane gets building insights 60x faster with Amazon Bedrock AgentCore
AWS Machine Learning Blog · 22 September 2026
Loading the full article…
This text was published by AWS Machine Learning Blog and written by Senthil Chinnaiyan. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Agents & Tools
All →- OpenAI's GPT-6 Astra can control Blender to create 3D models · 1 src
- Opinion: AI could help strawberry farmers forecast harvests and labor · 1 src
- SpaceXAI says Grok Bot reached 418,000 users, up 24% week over week · 1 src
- Rabbit launches OS3 agentic operating system for desktop and mobile · 1 src
- Paper2Agent turns research papers into interactive AI agents · 1 src
Comments
via GitHub Discussions