GitHub user geserdugarov added a comment to the discussion: [DISCUSS] New
Agentic lakehouse module & built-in AI agent gateway
I'm confused by the way Hudi is mentioned both as a source of tabular data and
as a source of data with embeddings.
### Scenario 1
Question:
> What was the total revenue by country last month?
The pipeline could be:
```text
user question
--> discover catalog/schema
--> generate SQL
--> execute through Trino
--> return a bounded aggregate/result set
--> formulate the answer
```
### Scenario 2
Question:
> Which customer complaints mention intermittent connection failures after an
> upgrade?
The pipeline could be:
```text
user question
--> create query embedding
--> vector or hybrid search
--> retrieve relevant documents/chunks
--> place them in the model context
--> formulate the answer
```
The two scenarios could be combined, but to start with, I propose **focusing on
Scenario 1**.
GitHub link:
https://github.com/apache/hudi/discussions/19264#discussioncomment-17711633
----
This is an automatically sent email for [email protected].
To unsubscribe, please send an email to: [email protected]
