A digital thread through your engineering data
Engineering knowledge sits in drives, workstations, spreadsheets and PLM, ERP and test exports. We gather it into one catalog with the context of each run attached, and Prism lets every team ask it questions and check the answer.
Data management
Scattered files become a searchable catalog
Most AI projects in engineering stall at the data, not the model. We do the preparatory work between your legacy records and the systems that will use them.
- Locate
- Find simulation runs, models and reports across drives, workstations and archives.
- Harvest
- Extract the metadata that makes a run reusable: why it was run, when, how, and what came out.
- Catalog
- Legacy files and new runs land in one searchable catalog, ready for analysis and for training surrogate models.
Curlscape Prism — Engineering Data & Knowledge
Prism puts every team on the same numbers. It brings engineering, quality and supply-chain data together under your own metric definitions and business rules, and shows the records behind every answer.
Prism screen recording: an Excel workbook of fleet reliability data is dropped in and loaded as queryable tables. Prism asks three clarifying questions about engine codes, the MTBF metric and repair cost, then drafts column descriptions, metric definitions and table joins for review.
Prism screen recording: asked about vessel downtime by root cause, Prism writes the SQL and returns a stacked bar chart with its columns and query shown. A follow-up ranks the fleet by MTBF, using the metric defined at import.
Prism screen recording: answers are pinned to a fleet reliability dashboard. Asking the board to add repair cost by component adds a new bar chart, and hovering any figure shows its values.
How it works:
- Bring your data in
- Spreadsheets and exports from PLM, ERP, test archives and quality systems load as queryable tables, merged headers and banner rows included.
- Teach it what the data means
- Your engineers record what each column means, the business rules, the metric definitions and how tables join. Every answer has to obey them.
- Ask in plain English
- Answers and charts are computed from the rows a query actually returned, with the query kept alongside, so any number can be checked.
- Govern the numbers
- Approved metrics are defined once and mean the same thing on every board, so design reviews start from agreed figures.
FAQ
Clearly answered
01What is a digital thread in engineering?
A digital thread links the data an engineered product generates (CAD, simulation runs, test results, quality and supply-chain records) so it can be found, traced and reused. Curlscape builds one by locating that data across drives and archives, extracting the metadata that makes each run reusable, and cataloguing it, then lets teams question it with Curlscape Prism.
02What data can Curlscape Prism work with?
Curlscape Prism loads spreadsheets and exports from PLM, ERP, test archives and quality systems as queryable tables, including messy sheets with merged headers and banner rows. Your engineers record what each column means, the business rules, the metric definitions and how tables join, and every answer has to obey them.
How Prism works03How do I know a Prism answer is correct?
Every Curlscape Prism answer is computed from the rows a query actually returned, and the query is kept alongside it, so any number can be checked against its source records. Approved metrics are defined once, so the same figure means the same thing on every board.