DataFlow-Harness: A Grounded Code-Agent Platform for Constructing Editable LLM Data Pipelines

Runming He, Zhen Hao Wong, Hao Liang, Zimo Meng, Chengyu Shen, Xiaochen Ma, Wentao Zhang
July, 2026
PDF Code arXiv

Abstract

DataFlow-Harness bridges the gap between code generated from natural language and persistent, editable data-pipeline artifacts. It combines procedural skills, live platform context exposed through MCP, and a synchronized visual DAG editor for grounded pipeline construction.

Type
Preprint
Publication
arXiv preprint arXiv:2607.16617
Xiaochen Ma
Xiaochen Ma
Ph.D. Student in Computer Science and Engineering

I work on data infrastructure and high-performance distributed systems for large-scale LLM data preparation, with a focus on Ray and Apache Spark.