← Back
AWS ML Blog

How CrossnoKaye grounds physical AI with Amazon Bedrock AgentCore

•16 min read•
#agents#bedrock#amazon
Level:Advanced
For:AI Engineers
✦TL;DR

CrossnoKaye built ATLAS and its agentic AI system AURA on Amazon Bedrock AgentCore to deliver physical AI across > 100 industrial facilities, leveraging native plant context and Bedrock’s code‑interpreter to run Python workloads in isolated microVM sandboxes. The platform stores > 22 trillion machine data points, ingests > 50 billion new readings daily, and runs > 1,000 live APC applications that boost throughput by 11–17 % and cut energy costs by ≈ 21 %. For engineers, this shows how tightly integrated Bedrock AgentCore can enable closed‑loop, data‑rich AI control without retrofitting plant hardware.

⚡ Key Takeaways

  • CrossnoKaye processes > 22 trillion data points and > 50 billion new readings per day using ATLAS.
  • AURA uses Amazon Bedrock AgentCore’s code‑interpreter to execute Python in AWS‑managed microVM sandboxes.
  • Production APC apps improve throughput 11–17 % while reducing facility energy costs by an average of 21 %.
  • Integration relies on a single security boundary that gives AURA native access to PLC/DCS, historians, and optimization tools via ATLAS.
  • The solution requires on‑prem ATLAS edge panels and AWS services (Bedrock AgentCore, Guardrails, microVMs); without this stack the agent cannot access full plant state.
💡 Why It Matters

Engineers can now deploy LLM‑driven agents that safely compute real‑time physics and statistics on massive OT streams, delivering measurable efficiency gains without costly hardware overhauls.

✅ Practical Steps

  1. Deploy ATLAS edge control panels on‑premises to ingest plant sensor streams into AWS.
  2. Configure AURA to invoke Amazon Bedrock AgentCore’s code‑interpreter for Python tasks within isolated microVM sandboxes.
  3. Enable Amazon Bedrock Guardrails and local edge safety controls to enforce security and safety boundaries for all agent actions.

Want the full story? Read the original article.

Read on AWS ML Blog ↗

More like this

Introducing workload recovery and expanded infrastructure checks on Amazon SageMaker HyperPod

AWS ML Blog•#deployment

Introducing Gemini 4 Neon

Google DeepMind•#llm

DaisyRose-27B: Pushing the Frontier of Post-Training Efficiency

Hugging Face Blog•#llm

Mathematical proof, powered by AI. And no hallucinations!

Amazon Science•#agents

EXPLORE AI NEWS

Daily hand-picked stories on LLMs, RAG, agents and production AI — curated for engineers who ship.

BROWSE NEWS

GET THE WEEKLY DIGEST

Join engineers getting the Monday signal-over-noise AI breakdown. No spam, unsubscribe anytime.

LEARN AI ENGINEERING

Curated courses, research papers, repos and tutorials built for engineers leveling up in AI.

START LEARNING