The field of natural language interfaces to data is moving towards more accurate and efficient text-to-SQL models, with a focus on improving the performance of large language models (LLMs) in real-world applications. Recent developments have highlighted the importance of high-quality training data, dataset alignment, and domain-specific knowledge in achieving state-of-the-art results. Noteworthy papers in this area include: LLMSQL, which introduces a revised and transformed version of the WikiSQL dataset designed for the LLM era, providing a clean and LLM-ready benchmark for text-to-SQL research. Retrieval and Augmentation of Domain Knowledge for Text-to-SQL Semantic Parsing, which proposes a systematic framework for associating structured domain statements with databases, demonstrating improved accuracy and practicality over existing approaches. Do LLMs Align with My Task, which studies the problem of dataset alignment for NL2SQL tasks and shows that structural alignment is a strong predictor of fine-tuning success. Agent Bain vs. Agent McKinsey, which introduces a new benchmark for text-to-SQL in the business domain, highlighting the gap between popular LLMs and the need for real-world business intelligence.