Engineering Knowledge Copilot for Bertrandt

Executive Summary

Challenge in One Sentence

Bertrandt is looking for an AI-powered Engineering Knowledge Copilot that collects internal engineering knowledge and makes it easily and reliably accessible to engineers during their daily work, in order to preserve expert know-how and support both experienced and newly hired engineers.

Why This Matters

Bertrandt is an engineering services company operating in vehicle development, with engineers in Germany and a growing international team that includes many young engineers fresh from university. Critical know-how is often not written down and is lost when experienced employees leave. Existing knowledge is scattered across PowerPoint, Word, PDF and Excel files, in people's heads and in customer process documents, and it is hard to keep current and hard to access in day-to-day work. A copilot that captures this knowledge and delivers reliable, source-based answers directly in the workflow would address this. A decisive factor is that capturing this knowledge must be as effortless as possible for engineers — ideally by simply speaking during their work — so that tacit know-how is recorded without extra burden.

Solution Space

We are looking for solutions that can provide:

• Capture scattered and tacit engineering knowledge with low effort — from existing documents and directly from employees, ideally by simply speaking (voice input)
• Knowledge is also available (in part) in the form of presentations, documents, standards, etc.
• There is also a large database of presentations from various conferences, at which OEMs presented their latest technologies.
• Provide a copilot / agent that answers engineering questions in daily work via a web interface or Microsoft Teams
• Deliver reliable, source-referenced answers with traceability to avoid hallucinations
• Support German and English, including for international teams
• Especially for the onboarding of new employees, particularly in other countries
• Ideally with access to CAD data to provide support using examples of existing components
• Build on existing products or reusable modules rather than a full greenfield build
• Focus the PoC on general, customer-independent engineering knowledge – design guidelines with a focus on metal and plastic parts and general legal homologation (type-approval) requirements – with (automotive) cockpit module development as the lead use case

The Venture Client project focuses on validating a working copilot in a controlled PoC before broader rollout and further extension.

Process (Overview)

This challenge offers solution providers the opportunity to validate their technology inside a leading automotive engineering services company and to co-shape a knowledge copilot with strong scaling potential.

Participants can gain:

• A financed PoC within a structured Venture Client framework
• Direct collaboration with Bertrandt engineers and the internal AI/IT team
• Access to real engineering use cases, sample documents and pilot users
• A clear pathway toward a long-term solution and buy-out or licensing
• A strong enterprise reference case in the automotive engineering domain

Who can participate?

Startups, scaleups and technology companies building AI copilots, RAG systems or agentic knowledge-management solutions for enterprise engineering environments, ideally with reusable core components and a European delivery footprint.