Accelerating Technical Insight with Scalable AI for Global Automotive Operations

Global automotive manufacturers depend on vast volumes of technical documentation to design, manufacture, and maintain complex vehicle systems. Akkodis partnered with a leading Italian multinational automotive enterprise to deliver a scalable, AI-driven solution that accelerates technical insight, improves measurement accuracy, and enables near real-time access to engineering data worldwide.

Customer Overview: Global Automotive Engineering at Scale

The customer is a distinguished Italian multinational enterprise specializing in the design, manufacturing, and commercialization of light, medium, and heavy-duty commercial vehicles, including trucks, vans, and buses. Supporting global operations requires precise, consistent interpretation of thousands of technical drawings across multiple revisions.

As the volume and complexity of engineering documentation increased, manual measurement validation became time-intensive and vulnerable to inconsistencies—creating the need for a more intelligent, automated approach.

Engineering precision at scale requires automation that can keep pace with global operations.

The Challenge: Extracting Accurate Measurements from Complex Drawings

The primary challenge was to develop an AI agent capable of extracting measurements from a large dataset of technical drawings using concise prompts. Each prompt needed to specify attributes such as the requested measurement, the component, and the referenced model.
In addition, the AI agent was required to:

• Detect potential measurement inconsistencies across drawing revisions
• Support multiple concurrent requests from technical users worldwide
• Operate efficiently on datasets consisting of thousands of drawings
• Scale automatically to meet fluctuating global demand

Ensuring accuracy, speed, and resilience was critical to supporting engineering teams across regions.

Accuracy and scalability were non-negotiable requirements for global engineering teams.

AWS-Powered AI Solution Designed for Scale and Security

Akkodis designed and implemented the solution using Amazon SageMaker AI as an IDE-like development environment. The architecture integrates several AWS services to enable intelligent automation and secure data handling:

• Amazon Bedrock for large language model (LLM) API calls
• Amazon Textract and Amazon Rekognition for OCR and visual data extraction
• Amazon Cognito for secure Single Sign-On (SSO) authentication
• Amazon S3, DynamoDB, Lambda, and API Gateway for scalable, serverless operations

The solution architecture is fully serverless, preventing persistent storage of model inputs and outputs and ensuring sensitive data is not shared with third-party providers.

Serverless architecture ensures scalability while maintaining strict data governance.

Performance, Resilience, and Measurable Impact

The AI-powered pipeline exceeded all customer-defined precision thresholds and reduced response times from minutes to seconds—enabling near real-time delivery of technical measurements.
Key performance metrics included:

• Request success rate (percentage of responded requests)
• Accuracy rate (percentage of correctly answered requests)

To ensure operational continuity, workload resilience was achieved through a multi-region AWS deployment, supporting uninterrupted access for global engineering teams.

The project was co-developed with the customer across all phases, following a collaborative R&D model focused on applying novel AI technologies to real-world engineering challenges.

Near real-time insights empower faster, more confident engineering decisions.

The Outcome: Intelligent Automation for Technical Documentation

This solution demonstrates how AI can enable organizations that rely on technical documentation—from structured text to complex engineering drawings—to automate management and review workflows.

By improving accuracy, consistency, and maintainability, AI-driven automation supports higher-quality engineering outputs while significantly reducing manual effort. The approach is broadly applicable across industries where precision and scalability are essential.

AI transforms technical documentation into a strategic operational asset.