AWS re:Invent 2025 - 4x faster workload modernization with agentic AI (MAM349)

Published: 1 hour ago (December 5, 2025 at 03:47 PM EST)

3 min read

Source: Dev.to

Overview

AWS re:Invent 2025 – 4x faster workload modernization with agentic AI (MAM349)

In this session, AWS Senior Solutions Architects Lakshmi Dasari and Vimala Pydi introduce AWS Transform, an agentic AI service that accelerates cloud migration and modernization by up to 4×. They note that 70 % of workloads still run in traditional data centers and typical migrations take 12–18 months. AWS Transform leverages specialized agents trained on 19 years of AWS expertise to automate discovery, planning, and migration phases.

The demo showcases a VMware migration with automated wave planning, network‑topology transformation, and VPC deployment. Vector Limited reported 20–34 % faster migration and 35 % TCO savings over five years. The session also references the AWS Well‑Architected Framework, which includes 57 questions across six pillars for ongoing optimization.

This article is auto‑generated from the original presentation content. Typos or minor inaccuracies may be present.

Migration Challenge: From Personal AI to Enterprise‑Scale Transformation

“I turned to Claude, our agentic AI, and created a personalized activity book for my kids during a road trip. That’s a small example of how agentic AI drives personal productivity. At AWS we aim to solve much bigger problems with this technology.”

Lakshmi Dasari (Senior Solutions Architect, AWS) and Vimala Pydi (Senior Migration Solutions Architect, Public Sector) introduce the session.

Generative AI Adoption

~80 % of enterprises will deploy generative AI‑enabled applications by 2026.
>15 % of work decisions are expected to be automated by agentic AI by 2028 (up from 0 %).
Human oversight remains a critical component of these autonomous decisions.

Technical Debt Landscape

70 % of workloads still run on traditional data centers and on‑premises technology.
>70 % of legacy IT software is older than 20 years, making it hard to maintain, brittle, and difficult to modernize.

Cloud Adoption Journey

Organizations move through various phases of cloud adoption. Real value is realized when they deliberately retire technical debt and adopt cloud‑native technologies, enabling true reinvention. Typical migrations can take 18 + months to complete.

Migration Methodology

A conventional migration process includes four phases:

Assessment – Build a business case, define ROI, and secure executive buy‑in.
Discovery – Deep dive into the current IT state, application lifecycles, and dependencies.
Planning – Determine migration strategies (rehost, retire, re‑platform, re‑architect) and design wave‑based migration plans.
Execution – Build the landing‑zone architecture in AWS and perform migration and modernization.

Scaling Migration with Agentic AI

Processing thousands of servers manually is impractical. Agentic AI, combined with large language models (LLMs) and graph neural networks, can:

Analyze millions of lines of code.
Extract business logic and map application dependencies.
Infer network topology and infrastructure relationships.
Generate the target infrastructure artifacts required for successful migration.

AWS Transform: Accelerating Migration with Specialized Agents

AWS Transform provides a suite of specialized agents that automate the end‑to‑end migration workflow—from discovery to landing‑zone creation—delivering up to 4× faster migrations and significant cost savings.