Raymond Cheng

### What is it? Assuming we do this https://github.com/opensource-observer/oso/issues/2194 it should be trivial.

Generic dlt + GraphQL factory

1

### What is it? Some GraphQL APIs support schema introspection https://graphql.org/learn/introspection/ This is something the developer has to turn on (we do for our API). If it exists, then we...

AlgoliaSearchList should have some better default results

2

### What is it? Before users enter a query we get a seemingly random set of projects back this should be a list of top queries or something

Meter on query complexity in the OSO API

1

### What is it? 2 queries against the OSO API may not have the same cost on Clickhouse. We should add some way to estimate query complexity

text2sql dataset + experiment

2

### What is it? we should have an MVP test suite to see how SQLAgent and SemanticAgent are performing

text2sql eval

1

### What is it? We want to do Dot-style evals, where we run the "correct" query on OSO and compares results.

SemanticAgent MVP

1

### What is it? After we finish SQLAgent, we should create text-to-semanticquery, which should perform better.

Epic: llmoso supports single-stage Information Retrieval

1

### What is it? Basic information retrieval is just asking questions that we think can be solved with a single query to OSO. In-scope examples: - Which projects have a...

Team workspace includes chat history

1

### What is it? Across team members

Team workspace includes saved queries

1

### What is it? To share across team