Question 1

What is ekai?

Accepted Answer

ekai is the semantic layer between your data cloud and your AI agents. It profiles your data in place, asks the people who hold the meaning, turns what they know into governed semantic models and production dbt pipelines, and verifies every definition against your warehouse before anything ships. Months of consulting, done in hours.

Question 2

Who is ekai for?

Accepted Answer

Data teams who own the warehouse, the domain experts who hold the definitions, and the business users who ask the questions. Data teams get semantic models and dbt pipelines without a months-long modeling engagement. Experts answer in plain language, and only about what the data cannot explain on its own. Business users ask in chat and get answers that carry the same definitions everywhere. Nobody needs to write YAML to take part.

Question 3

How does ekai work?

Accepted Answer

In the order disciplined data modeling has always worked: meaning first, then structure, then build. ekai profiles your data before it asks anything, so the exchange with your specialists is never a generic checklist. From those profiles and their answers it infers entities, keys, and relationships, and shows the evidence behind every inference for a human to approve. Then dbt models, tests, documentation, glossary, metrics, and lineage are generated together, executed inside your warehouse, and reconciled against the business facts your experts stated up front.

Question 4

How do I know the model is right?

Accepted Answer

Nothing ships unchecked. Your experts state known truths up front, such as revenue matching the finance close, and every model has to reproduce them before it can publish. Schema tests, validation rules, and documentation are generated with the model rather than bolted on afterwards, and every inferred relationship, definition, and metric is reviewed by your people, with ekai showing its evidence and taking correction.

Question 5

Does my data ever leave my environment?

Accepted Answer

No. ekai installs as a Snowflake Native App inside your own account, or as dedicated SaaS on AWS, Azure, or GCP. Connections are read-only and profiling happens in place, inside your warehouse: nothing crosses the boundary, not rows, not profiles, not metadata. Model traffic can route privately through your own Vertex AI or Azure OpenAI endpoints, so it follows the cloud agreements you already have. More detail is on our security page.

Question 6

Which platforms does ekai support?

Accepted Answer

Snowflake, Databricks, BigQuery, Azure Synapse, and PostgreSQL. Data connection, ingestion, and artifact generation work across all five; native semantic agent integration covers every one of them except Azure Synapse today. Finished models publish to Snowflake Intelligence and Cortex Agents, Databricks Genie, and your data catalogs.

Question 7

What do I actually get, and can I keep it?

Accepted Answer

Open, portable artifacts: standard dbt projects, YAML semantic views, JSON catalogs, business glossaries, metric definitions, and lineage. Every artifact is versioned to your own GitHub, GitLab, or Bitbucket, under your review process and your audit trail. ekai has no warehouse to lock you into and no incentive to model toward one vendor's stack, so use ekai to build these and keep them for good.

Question 8

How long does it take to get value?

Accepted Answer

Hours rather than months. Documenting what your people already know the manual way runs months and six figures, and by delivery the business has moved on. A first engagement with ekai is deliberately small: bring one domain and three business questions, and leave with a reviewed model, the pipeline that builds it, and the tests that prove it.

AI-ready context shouldn’t take months.

30 minutes with our team.

Manual extraction is slow.

Unverified AI is a liability.

Governance stalls before production.