{"slug": "show-hn-relativedb-oss-query-engine-for-relational-foundation-models", "title": "Show HN: RelativeDB – OSS query engine for relational foundation models", "summary": "RelativeDB, an open source query engine built around the Relational Transformer architecture from Stanford's Tabular and Relational (STAR) Project, can outperform XGBoost on some tasks with minimal labeled context and no manual feature engineering. The engine allows users to query relational foundation models using SQL-like syntax for predictions such as customer churn, demand forecasting, fraud detection, and recommendations.", "body_md": "Structured data could be the next big frontier for foundation models.\n\nRelativeDB is an open source query engine built around the Relational Transformer architecture, developed by Stanford’s Tabular and Relational (STAR) Project. These models can adapt to new databases and prediction tasks using relatively little labeled context.\n\nWith some of my experiments, the results have been remarkable. On some tasks, RelativeDB can outperform XGBoost with only a small amount of relational context and no manual feature engineering.\n\nSo RelativeDB turns relational foundation models into something you can query (like this customer churn query):\n\n``` PREDICT NOT EXISTS(orders.*) OVER (90 DAYS FOLLOWING) FROM customers ```\n\nThe engine finds the relevant context across the related tables and returns a prediction. The goal is to make questions like churn, demand forecasting, fraud detection, and recommendations feel more like database queries than bespoke ML projects.\n\nLonger term, I think the optimizer should be able to choose between relational foundation models, fine-tuned models, and conventional approaches such as XGBoost depending on the query (how much graph-like context is needed vs just tabular features). Bridging ML to LLMs in a sane way.\n\nI think there are markets for quantitive finance and utilizing emerging context graphs.\n\nThe project is early, and I would especially appreciate feedback on the query language and execution model.\n\nRelativeDB: [https://github.com/RelativeDB/RelQL](https://github.com/RelativeDB/RelQL)\nStanford RT-J: [https://star-project.stanford.edu/rt-j/](https://star-project.stanford.edu/rt-j/)\n\nComments URL: [https://news.ycombinator.com/item?id=49063842](https://news.ycombinator.com/item?id=49063842)\n\nPoints: 2\n\n# Comments: 0", "url": "https://wpnews.pro/news/show-hn-relativedb-oss-query-engine-for-relational-foundation-models", "canonical_source": "https://github.com/RelativeDB/RelQL", "published_at": "2026-07-27 00:32:52+00:00", "updated_at": "2026-07-27 00:52:17.566549+00:00", "lang": "en", "topics": ["machine-learning", "ai-products", "ai-research", "ai-infrastructure"], "entities": ["RelativeDB", "Stanford Tabular and Relational (STAR) Project", "XGBoost", "Relational Transformer"], "alternates": {"html": "https://wpnews.pro/news/show-hn-relativedb-oss-query-engine-for-relational-foundation-models", "markdown": "https://wpnews.pro/news/show-hn-relativedb-oss-query-engine-for-relational-foundation-models.md", "text": "https://wpnews.pro/news/show-hn-relativedb-oss-query-engine-for-relational-foundation-models.txt", "jsonld": "https://wpnews.pro/news/show-hn-relativedb-oss-query-engine-for-relational-foundation-models.jsonld"}}