... Hunters! I'm building SlothDB an open-source embedded database that lets you query files directly with SQL. No server, no import step, no dependencies. You literally just do this: SELECT * FROM 'sales.csv'; SELECT * FROM read_parquet('events.parquet'); SELECT * FROM 'report.xlsx'; It supports CSV, Parquet, JSON, Excel, Avro, Arrow, and SQLite all built into the engine. Before we launch, I'd love to hear from you: 1. What file format do you deal with the most at work? 2. What ... ... working with data files? 3. Would GPU-accelerated queries (CUDA + Metal) matter for your use case? We're open source (MIT) and you can try it right now: curl -fsSL https://raw.githubusercontent.co... | bash Or just: pip install slothdb