This paper introduces Agentics, a functional agentic AI framework for building LLM-based structured data workflow pipelines. Designed for both research and practical applications, Agentics offers a new data-centric paradigm in which agents are embedded within data types, enabling logical transduction between structured states. This design shifts the focus toward principled data modeling, providing a declarative language where data types are directly exposed to large language models and the data values are composed through transductions between input and output types. We present a range of structured data workflow tasks and empirical evidence demonstrating the effectiveness of this approach, including data wrangling, text-to-SQL semantic parsing, and domain-specific multiple-choice question answering, and data-driven scientific discovery tasks.
@article{arxiv.2508.15610,
title = {Transduction is All You Need for Structured Data Workflows},
author = {Alfio Gliozzo and Naweed Khan and Christodoulos Constantinides and Nandana Mihindukulasooriya and Nahuel Defosse and Gaetano Rossiello and Junkyu Lee},
journal= {arXiv preprint arXiv:2508.15610},
year = {2026}
}