SemPiper: Interactive Code Synthesis for Semantic Operators in Machine Learning Pipelines
Machine learning (ML) pipelines require extensive data prepara tion, feature engineering, and integration across heterogeneous sources, making them tedious and error-prone to develop. While large language models (LLMs) have recently shown promise for assisting programming tasks, chat-based interfaces provide lim ited control over pipeline behavior and often produce code that is difficult to optimize or integrate into production systems. We demonstrate SemPipes, a novel programming model that extends ML pipelines with declarative, LLM-powered semantic data op erators. SemPipes allows developers to specify high-level natural language instructions for data-centric operations, while seamlessly combining these operators with arbitrary Python code from stan dard data sciencelibraries. For the semanticoperators,it synthesizes specialized implementations at pipeline training time, conditioned on dataset characteristics and pipeline context, enabling the flexi ble yet controlled integration of LLM capabilities. We demonstrate SemPipes through SemPiper, an interactive interface that visual izes computational graphs of the pipelines, synthesized operator implementations, and optimization trajectories produced by an evo lutionary search procedure. Attendees can explore three end-to-end scenarios, modify pipelines, inspect generated code, and observe how semantic operators are synthesized and iteratively optimized. The demonstration highlights how declarative semantic operators enable controllable, optimizable, and practical integration of LLMs into ML pipeline development.
