## Metadata - Author: [[jason-ganz|Jason Ganz]] - Full Title:: Semantic Layer as the Data Interface for LLMs - Category:: #🗞️Articles - Document Tags:: [[Foundation models|Foundation Models]], - URL:: https://roundup.getdbt.com/p/semantic-layer-as-the-data-interface - Finished date:: [[2024-01-17]] ## Highlights > there are now multiple companies working to help you do that in production. There is [Delphi](https://www.delphihq.com/), who I linked to in my prior writeup. There’s [Dot](https://www.getdot.ai/), a data driven chat interface. There’s [Seek AI](https://www.seek.ai/), who just achieved [best in class SQL generation benchmarks](https://twitter.com/sarahrnagy/status/1722036279045460291). All three of these companies [directly](https://www.delphihq.com/integrations/dbt) [integrate](https://blog.getdot.ai/chatgpt-on-dbt-semantic-layer-bring-analytics-self-service-to-the-next-level-with-dot-f73fe970bf42) [with](https://www.seek.ai/blog/seek-ai-partners-with-dbt-labs-to-enhance-customers-generative-ai-for-their-data) the dbt Semantic Layer. ([View Highlight](https://read.readwise.io/read/01hmazyn6xdasepy2jhzvtkftj)) > if you are building on top of a modern data stack, you are building the foundation for an AI-powered analytics workflow to drive your business forward. ([View Highlight](https://read.readwise.io/read/01hmazz3jba8kyw70sjmfkj189))