DEMA Engineering Company, a leading supplier of chemical dispensing systems and industrial valves, needed a way to complete Design Failure Mode and Effects Analysis faster without sacrificing quality. Each new product triggered 6-8 meetings with up to 10 engineers, requiring them to manually reconstruct risk assessments that already existed in historical documentation.
SoftSnow built an AI Risk Assessment Agent that systematizes decades of institutional knowledge. The agent pulls from product-specific databases stored in SharePoint, synced to Cassidy AI, a work automation platform that executes multi-step workflows, then processes complex Excel files, and delivers comprehensive failure analyses that match DEMA's existing DFMEA template format. Engineers provide three inputs - product name, product type, and a component file - and the agent returns a populated spreadsheet with potential failure modes, effects, causes, and estimated risk priority numbers.
The result:
- 100% institutional knowledge coverage in every assessment.
- Engineering teams now arrive at meetings with pre-populated analysis instead of blank templates, shifting conversations from "What could go wrong?" to "Is this assessment accurate?"
- Product cycles accelerated.
- Safety reviews no longer bottleneck development.
The system improves continuously as new products and failure modes get added to the knowledge base.
"We're not just saving time," the operations lead reflected during the demo. "We're designing a system that gets smarter with every product we develop."
Engineering teams accumulate tremendous expertise through product development. That knowledge represents a competitive advantage if you can systematize it.
Want to explore how systematic intelligence could expand your team's capacity? Let's talk.
