This dataset contains 10,000+ complex SQL-style analytical questions
mapped to MongoDB queries and aggregation pipelines.
Multiple schemas
$group, $sum, $avg, $lookup
Nested documents
Long analytical questions
Fine-tuning small LLMs (Qwen, Mistral, LLaMA 3B)
Text-to-Mongo query generation
Data analytics agents
