AI Hackathon for Manufacturing

Turn SOP's into an AI Assistant

Co-hosted by Xebia & AWS
March 26, 2026 | Xebia Eindhoven

Manufacturing teams deal with complex Standard Operating Procedures every day: Maintenance steps, safety rules, quality checks, escalation paths. These documents are essential, but often hard to use in practice.

During this one-day hackathon, you will learn how to convert SOPs into a working AI assistant that can answer questions, guide operators, and surface the right information at the right moment using AWS-native AI services.

And the best thing is that you will build, test, and refine the solution yourself, with guidance from Xebia and AWS specialists.

What You Will Build

You will work on a shared use case used throughout the day:

An AI SOP Assistant for Manufacturing, capable of:

• Interpreting SOP documentation (PDFs, manuals, procedures)
• Answering operator questions in natural language
• Returning precise, source-based answers using Retrieval-Augmented Generation (RAG)
• Running on AWS using Amazon Bedrock and supporting services

The emphasis is on architecture choices, data preparation, prompt design, and reliability (not just getting a chatbot to respond).

What You Will Learn

By the end of the day, you will understand:

• How to structure SOP content so AI can reason over it reliably
• How RAG works in practice for manufacturing documentation
• How to design prompts that produce consistent, usable answers
• How to reduce hallucinations and keep responses grounded in source data
• How to move from a static document set to an interactive assistant
• How AWS services (including Amazon Bedrock) fit together in a production-ready setup

You will leave with a clear blueprint you can apply inside your own organization.

Optional: Apply It to Your Own Use Case

If you have a relevant dataset or challenge, you can apply the same approach to:

• Maintenance instructions or failure logs
• Quality or inspection procedures
• Safety documentation
• Operational or compliance manuals

Xebia and AWS experts will support teams in adapting the architecture to their context.

Who Should Attend

Teams of 2–4 participants, typically including:

• Software engineers or data engineers
• Manufacturing IT or digital leads
• Innovation managers or product owners

No prior AWS or GenAI experience is required, experiences with cloud services is useful.

Agenda

Practical Details

• Location: Xebia Eindhoven, Willemstraat 1K
• Hardware: Bring your own laptop
• AWS Environment: Fully provisioned sandbox environments provided
• Data: Optional anonymized SOPs or documents if bringing your own case

Registration

Participation is limited to keep teams small and support hands-on learning.