This presentation, hosted by Quinlan's Lab for Applied Artificial Intelligence, focuses on the University of Chicago’s Multi–Millennia Climate Reanalysis through Chinese Historical Records (MMCRCHR) Project as a primary example of how scholars can develop open-source, modular, scalable, extensible and reusable Agentic AI infrastructures that combine modern text analytical models and frameworks with scientific data sources. The initial prototype for this new system is designed to provide a reproducible and sustainable research platform to begin to provide answers to the following question: “How do representations of climate change in the 3000-year Chinese historical record align with modern climate data?” MMCRCHR was designed with human oversight built into every step in the pipeline in an attempt to identify and counter the relatively high error rates produced by current transformer-based models/systems (“hallucinations”), providing both a summary of the results as a response to a user’s query and all relevant underlying prima facie evidence and sources used to produce the summary back to the user: citations from textual sources plus relevant datasets returned from scientific data archives via API. The current backend employs an agent LLM that retrieves passages from the 40 million characters in the Chinese Official Histories 正史 (24 extensive compendia dating from 90 BCE to 1912 CE, with each passage/chunk dated in preprocessing); we are currently expanding the framewotk to include a multilingual corpus of hundreds of modern and premodern secondary sources on climate change and its effects to serve as our secondary large textual data source, and greatly increasing the number of scientific datasets available to the system. Embeddings leveraged by transformer-based agentic infrastructures have recently enabled a wealth of new approaches and have come to be widely viewed as “the future of academic research,” but ethics, reliability, reproducibility and sustainability remain significant issues that we as a community need to address; we see the MMCRCHR as a potentially groundbreaking step forward in these areas.