Daikin Industries has selected Outerport as its core platform for extracting structured data from complex engineering documents, drawings and diagrams. The Outerport platform has been fully deployed to streamline Daikin's design engineering operations, paving the way for the widespread build-out of AI agents across their engineering organization.

The bottleneck for AI agents in manufacturing: unstructured data

While the world of software development is experiencing rapid automation thanks to large-language models and autonomous AI agents (like Claude Code), the manufacturing industry lags in adoption.

In design engineering, decades of highly valuable design assets and technical know-how are trapped in unstructured formats, like paper documents, PDFs, diagrams and drawings. Because AI models cannot effectively read, search, or parse this complex visual data, manufacturers have been unable to use AI agents to automate their core design processes.

To solve this problem, Outerport develops a platform that bridges the gap between engineering documents and AI agents.

By combining proprietary computer vision technology with multi-modal large-language models (MLLMs), Outerport can convert unstructured visual data (like technical drawings and diagrams) into AI-ready formats (such as JSON). This high-fidelity document extraction allows AI agents to process, analyze, and reason over decades of engineering IP.

Built by a team of former research engineers who led AI development at major US tech giants such as NVIDIA and Meta Platforms, Outerport was selected by Daikin for its leading extraction accuracy and its ability to meet information security standards required for high-security manufacturers.

Choosing outerport for the future of design engineering

Daikin Industries recognized that to scale their engineering capabilities, they needed a purpose-built solution capable of turning unstructured data into assets for AI agents.

"Converting unstructured data into structured data has been a long-standing challenge and a major area of technical interest for us," said Shohei Hido, executive engineer at the Technology Innovation Center, Daikin Industries. "We had conducted internal trials on our technical documents, but we acutely felt the difficulty of achieving the accuracy required for actual business operations. Upon evaluating Outerport's technology, we were able to achieve a dramatic leap in parsing accuracy compared to conventional approaches. Beyond their technical capability, we highly value their agile development process and ability to customize the platform to our specific needs. We look forward to continuing working with Outerport to create entirely new design engineering processes."

Towards an AI-driven design engineering stack

Outerport's mission is to accelerate the speed of new hardware product development cycles for manufacturers through an "AI-driven engineering system" that can perform "design-for-X" checks to catch downstream issues in the engineering process as early as the design phase.

Moving forward, Outerport's goal is to continue to strengthen the competitiveness of the manufacturing sectors by focusing on three core pillars:

• Document extraction: Converting diagrams and drawings found in engineering documents into LLM-ready formats
•  AI Agent Systems: Developing a platform to allow companies to in-house build AI agents that leverage the structured data
• Domain Expertise: Applying specialized knowledge focused on manufacturing and engineering