AIBuildAI: An AI Agent for Automatically Building AI Models

arXiv cs.AI / 4/17/2026

📰 NewsIdeas & Deep AnalysisTools & Practical UsageModels & Research

共有:

Key Points

The paper introduces AIBuildAI, an AI agent that automatically builds AI models from a task description and training data, targeting the end-to-end lifecycle that is still labor-intensive for humans.
AIBuildAI uses a hierarchical agent setup where a manager coordinates three LLM-based sub-agents: a designer (modeling strategy), a coder (implementation and debugging), and a tuner (training and performance optimization).
Unlike prior AutoML approaches that mainly focus on limited steps such as hyperparameter tuning or model selection within preset spaces, AIBuildAI aims to automate the broader development process including design, coding, training, and evaluation.
In evaluations on the MLE-Bench benchmark of Kaggle-style tasks across multiple modalities (visual, textual, time-series, and tabular), AIBuildAI ranks first with a 63.1% medal rate and performance comparable to experienced AI engineers.
The results suggest hierarchical multi-agent systems could make AI development more broadly accessible by reducing the need for expert human intervention.

Abstract

AI models underpin modern intelligent systems, driving advances across science, medicine, finance, and technology. Yet developing high-performing AI models remains a labor-intensive process that requires expert practitioners to iteratively design architectures, engineer representations, implement training pipelines and refine approaches through empirical evaluation. Existing AutoML methods partially alleviate this burden but remain limited to narrow aspects such as hyperparameter optimization and model selection within predefined search spaces, leaving the full development lifecycle largely dependent on human expertise. To address this gap, we introduce AIBuildAI, an AI agent that automatically builds AI models from a task description and training data. AIBuildAI adopts a hierarchical agent architecture in which a manager agent coordinates three specialized sub-agents: a designer for modeling strategy, a coder for implementation and debugging, and a tuner for training and performance optimization. Each sub-agent is itself a large language model (LLM) based agent capable of multi-step reasoning and tool use, enabling end-to-end automation of the AI model development process that goes beyond the scope of existing AutoML approaches. We evaluate AIBuildAI on MLE-Bench, a benchmark of realistic Kaggle-style AI development tasks spanning visual, textual, time-series and tabular modalities. AIBuildAI ranks first on MLE-Bench with a medal rate of 63.1%, outperforming all existing baseline methods and matching the capability of highly experienced AI engineers. These results demonstrate that hierarchical agent systems can automate the full AI model development process from task specification to deployable model, suggesting a pathway toward broadly accessible AI development with minimal human intervention.