Building Tendos AI: How an Agent Swarm Turns Construction Emails into Quotes

The podcast highlights the inefficiencies in the construction industry's back-office processes, particularly within the tendering chain. Issues such as slow communication, manual workflows, and the need for complex coordination among various stakeholders result in delays and errors. To address these challenges, a new system called Tendos has been developed, which utilizes AI-powered automation to streamline administrative and planning tasks. The solution is designed to categorize, prioritize, and process customer requestsranging from emails and PDFs to large construction documentsusing large language models to extract relevant information, identify product needs, and generate accurate offers.

AI plays a key role in analyzing unstructured data and supporting decision-making, but the system still relies on human expertise for final approval in complex or customized cases. The product was developed iteratively, beginning with a focus on specific product types before expanding into more comprehensive construction workflows. It integrates with legacy systems like CRM and SAP, while supporting dynamic, context-aware processing. This integration aims to enhance efficiency, accuracy, and transparency across the entire construction value chain.