GHELIA AutoDeck Ver.1.1 adds AI chat knowledge creation

GHELIA AutoDeck is a platform that accumulates design and development knowledge—documents, 3D CAD models, simulation data, and images—and uses a multimodal LLM to generate explanatory text for unstructured data. The Ver.1.1 update strengthens both knowledge accumulation and search/utilization.

The new knowledge workspace feature lets an LLM converse with users in chat form to create knowledge, asking questions about template selection and data interpretation, and allowing additional file submissions. A knowledge template feature defines chapter structures such as purpose, observation, consideration, and conclusion, and sets conditions for LLM questions during creation; the LLM also automatically reviews content against acceptance criteria and can prompt rewrites. Templates can be created per process, department, or project.

For search, the update adds menus for related case search, cause investigation, tabular comparison, bullet points, and summaries, with speed-priority and accuracy-priority modes. Previously, search displayed a list of matching knowledge; now it cross-searches multiple entries and the LLM presents commonalities and abstracted solutions. Ghelia positions GAD as a "knowledge layer" above existing data management systems, aiming for knowledge to be used across processes by design, sales, and management.