How to move enterprise AI from pilot to production: AWS and Classmethod sign 5-year SCA

A five-year Strategic Collaboration Agreement between AWS and Classmethod sets a reusable path from AI pilots to production across manufacturing, retail and financial services.

Enterprises are stuck between two realities: AI saves individual employees time, but rarely transforms the business end to end. Legacy applications, fragmented data and security gaps block scale. On October 28, 2026, AWS and Classmethod, an AWS Premier Tier Services Partner in Japan, signed a five-year Strategic Collaboration Agreement (SCA) to close that gap with a repeatable model called Client Zero. Softprom is the official partner of AWS.

What was announced

Under the SCA, Classmethod will combine AWS services with its own in-production experience to deliver AI solutions across five focus areas: legacy modernization and security, managed operations, production AI for key industries, the Client Zero adoption model, and a new AI Experience Center targeted to open on November 2, 2026. The first-year target is 30 customers running industry AI solutions in production, starting with manufacturing predictive maintenance and know-how capture.

Classmethod reports that internal AI-driven development has increased coding productivity fivefold and overall project productivity twofold. Since 2011, Classmethod has supported more than 5,600 companies and helped build and manage more than 40,000 AWS accounts.

The most impactful partners are not just telling their customers how to transform with AI, they are doing it themselves. AWS and Classmethod are proving that model, extending all the way to Physical AI.

Why this matters

For CIOs, CISOs and procurement leaders in regulated industries, the agreement signals a shift from isolated AI pilots to governed, production-grade deployments. The combination of AWS Security Hub, Amazon GuardDuty and AWS Security Agent with modernization via AWS Transform addresses the growing attack surface of unmaintained legacy systems. Physical AI, backed by Amazon SageMaker AI and AWS IoT Greengrass, opens a direct path from analytics to autonomous action in factories, warehouses and critical infrastructure.

Technical details

• Legacy modernization: AWS Transform plus Classmethod migration services for applications, databases and data platforms.
• Security stack: AWS Security Hub, Amazon GuardDuty and AWS Security Agent for continuous monitoring, reviews and penetration testing.
• Managed operations: Amazon Bedrock and AWS DevOps Agent automate incident investigation, troubleshooting and root-cause analysis.
• Production AI: Amazon Bedrock with agentic capabilities for manufacturing, retail and financial services.
• Physical AI: simulation-based training, model refinement in Amazon SageMaker AI and edge inference via AWS IoT Greengrass.
• Internal stack: Classmethod runs on Amazon Bedrock, Kiro and Amazon Quick as its Client Zero environment.

FAQ

What is the Client Zero model?

Client Zero means the partner adopts each new AWS capability inside its own business first, codifies governance and patterns, then turns the result into a repeatable offering so customers deploy a proven approach instead of a first attempt.

When does the AI Experience Center open?

The AI Experience Center is targeted to launch on November 2, 2026. It will help customers design, prototype and validate AI use cases through to production architecture, governance, security and operations.

How does Physical AI differ from traditional AI projects?

Physical AI combines perception, reasoning and action on real equipment. Models trained in Amazon SageMaker AI are deployed to the edge with AWS IoT Greengrass, so robots and sensors can act in real time while feeding data back to improve the models.

Softprom and Amazon Web Services

Softprom is the official partner of Amazon Web Services. Our team helps enterprises assess AWS-based AI architectures, modernize legacy workloads and operationalize security controls.

Prepared by the Softprom editorial team based on vendor materials (Softprom DistriFlow). Source: original article.