← Back
AWS ML Blog

Scaling cloud migrations with agentic AI on Amazon Bedrock AgentCore

•20 min read•
#agents#bedrock#amazon
Level:Intermediate
For:Cloud Migration Engineers
✦TL;DR

Amazon Bedrock AgentCore powers a new multi‑agent framework that AWS Professional Services uses to automate end‑to‑end enterprise cloud migrations. The system deploys purpose‑built agents that perform discovery, generate infrastructure‑as‑code, enforce portfolio governance, and manage post‑migration operations, cutting the time spent on IaC development. By chaining agents through AgentCore’s orchestration layer, the solution scales to large, heterogeneous workloads while keeping governance and compliance in the loop. The approach demonstrates how agentic AI can replace manual migration scripts with a reusable, policy‑driven workflow.

⚡ Key Takeaways

  • The core model is Amazon Bedrock AgentCore, which orchestrates multiple specialized agents.
  • Agents are defined via declarative YAML configs that reference Bedrock LLMs and AWS SDK actions.
  • The framework trades off a modest increase in orchestration latency for a significant reduction in manual IaC authoring effort.
  • Integration is achieved by invoking agents through the Bedrock API endpoint, passing CloudFormation or Terraform templates as payloads.
  • The solution requires prior IAM permissions for CloudFormation, S3, and AWS Config; lacking these will halt agent execution.
  • WhyItMatters: Engineers building production migrations can replace ad‑hoc scripting with a repeatable, policy‑driven agent pipeline, accelerating deployment cycles and reducing human error in IaC.
  • TechnicalLevel: Intermediate
  • TargetAudience: Cloud Migration Engineers
  • PracticalSteps:
  • Create an AgentCore project in the AWS console and upload your migration spec YAML file.
  • Grant the project IAM roles that allow CloudFormation stack creation and AWS Config read/write.
  • Trigger the migration agent via the Bedrock API using the `InvokeAgent` action, passing the target account ID and region.
  • ToolsMentioned: Amazon Bed

🔧 Tools & Libraries

Amazon Bed
💡 Why It Matters

Engineers building production migrations can replace ad‑hoc scripting with a repeatable, policy‑driven agent pipeline, accelerating deployment cycles and reducing human error in IaC.

✅ Practical Steps

  1. Create an AgentCore project in the AWS console and upload your migration spec YAML file.
  2. Grant the project IAM roles that allow CloudFormation stack creation and AWS Config read/write.
  3. Trigger the migration agent via the Bedrock API using the `InvokeAgent` action, passing the target account ID and region.

Want the full story? Read the original article.

Read on AWS ML Blog ↗

More like this

Record, train, and deploy from one place with Strands Agents, LeRobot, and Hugging Face Storage Buckets

Hugging Face Blog•#agents

Add secure Web Search to Claude Desktop with Amazon Bedrock AgentCore

AWS ML Blog•#bedrock

From Training to Production, NVIDIA and CoreWeave Close the Loop on Agentic AI

NVIDIA Blog•#agents

Local Agentic AI Workflows with Hermes + Ollama

Machine Learning Mastery•#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