Saved in:
Bibliographic Details
Main Authors: Lu, Pengrui, Xu, Bingyu, Zhang, Wenjun, Hua, Shengjia, Gao, Xuanjian, Ge, Ranxiang, Ye, Lyumanshan, Wu, Linxuan, Li, Yiran, Yu, Junfei Fish, Zhang, Yibo, Li, Ruixin, Li, Manxiang, Han, Xiao, Zhou, Xiaocong, Chi, Guangyao, Chen, Zisheng, Chen, Kaishen, Wang, Kun, Xu, Qihua, Meng, Fengyue, Ni, Yuchen, Li, Jiajun, Liu, Jinxiu, Zhang, Danfeng, Zhao, Jingru, Liu, Pengfei
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.12162
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917406094393344
author Lu, Pengrui
Xu, Bingyu
Zhang, Wenjun
Hua, Shengjia
Gao, Xuanjian
Ge, Ranxiang
Ye, Lyumanshan
Wu, Linxuan
Li, Yiran
Yu, Junfei Fish
Zhang, Yibo
Li, Ruixin
Li, Manxiang
Han, Xiao
Zhou, Xiaocong
Chi, Guangyao
Chen, Zisheng
Chen, Kaishen
Wang, Kun
Xu, Qihua
Meng, Fengyue
Ni, Yuchen
Li, Jiajun
Liu, Jinxiu
Zhang, Danfeng
Zhao, Jingru
Liu, Pengfei
author_facet Lu, Pengrui
Xu, Bingyu
Zhang, Wenjun
Hua, Shengjia
Gao, Xuanjian
Ge, Ranxiang
Ye, Lyumanshan
Wu, Linxuan
Li, Yiran
Yu, Junfei Fish
Zhang, Yibo
Li, Ruixin
Li, Manxiang
Han, Xiao
Zhou, Xiaocong
Chi, Guangyao
Chen, Zisheng
Chen, Kaishen
Wang, Kun
Xu, Qihua
Meng, Fengyue
Ni, Yuchen
Li, Jiajun
Liu, Jinxiu
Zhang, Danfeng
Zhao, Jingru
Liu, Pengfei
contents The rapid deployment of AI agents in commercial settings has outpaced the development of evaluation methodologies that reflect production realities. Existing benchmarks measure agent capabilities through retrospectively curated tasks with well-specified requirements and deterministic metrics -- conditions that diverge fundamentally from production environments where requirements contain implicit constraints, inputs are heterogeneous multi-modal documents with information fragmented across sources, tasks demand undeclared domain expertise, outputs are long-horizon professional deliverables, and success is judged by domain experts whose standards evolve over time. We present AlphaEval, a production-grounded benchmark of 94 tasks sourced from seven companies deploying AI agents in their core business, spanning six O*NET (Occupational Information Network) domains. Unlike model-centric benchmarks, AlphaEval evaluates complete agent products -- Claude Code, Codex, etc. -- as commercial systems, capturing performance variations invisible to model-level evaluation. Our evaluation framework covers multiple paradigms (LLM-as-a-Judge, reference-driven metrics, formal verification, rubric-based assessment, automated UI testing, etc.), with individual domains composing multiple paradigms. Beyond the benchmark itself, we contribute a requirement-to-benchmark construction framework -- a systematic methodology that transforms authentic production requirements into executable evaluation tasks in minimal time. This framework standardizes the entire pipeline from requirement to evaluation, providing a reproducible, modular process that any organization can adopt to construct production-grounded benchmarks for their own domains.
format Preprint
id arxiv_https___arxiv_org_abs_2604_12162
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle AlphaEval: Evaluating Agents in Production
Lu, Pengrui
Xu, Bingyu
Zhang, Wenjun
Hua, Shengjia
Gao, Xuanjian
Ge, Ranxiang
Ye, Lyumanshan
Wu, Linxuan
Li, Yiran
Yu, Junfei Fish
Zhang, Yibo
Li, Ruixin
Li, Manxiang
Han, Xiao
Zhou, Xiaocong
Chi, Guangyao
Chen, Zisheng
Chen, Kaishen
Wang, Kun
Xu, Qihua
Meng, Fengyue
Ni, Yuchen
Li, Jiajun
Liu, Jinxiu
Zhang, Danfeng
Zhao, Jingru
Liu, Pengfei
Computation and Language
The rapid deployment of AI agents in commercial settings has outpaced the development of evaluation methodologies that reflect production realities. Existing benchmarks measure agent capabilities through retrospectively curated tasks with well-specified requirements and deterministic metrics -- conditions that diverge fundamentally from production environments where requirements contain implicit constraints, inputs are heterogeneous multi-modal documents with information fragmented across sources, tasks demand undeclared domain expertise, outputs are long-horizon professional deliverables, and success is judged by domain experts whose standards evolve over time. We present AlphaEval, a production-grounded benchmark of 94 tasks sourced from seven companies deploying AI agents in their core business, spanning six O*NET (Occupational Information Network) domains. Unlike model-centric benchmarks, AlphaEval evaluates complete agent products -- Claude Code, Codex, etc. -- as commercial systems, capturing performance variations invisible to model-level evaluation. Our evaluation framework covers multiple paradigms (LLM-as-a-Judge, reference-driven metrics, formal verification, rubric-based assessment, automated UI testing, etc.), with individual domains composing multiple paradigms. Beyond the benchmark itself, we contribute a requirement-to-benchmark construction framework -- a systematic methodology that transforms authentic production requirements into executable evaluation tasks in minimal time. This framework standardizes the entire pipeline from requirement to evaluation, providing a reproducible, modular process that any organization can adopt to construct production-grounded benchmarks for their own domains.
title AlphaEval: Evaluating Agents in Production
topic Computation and Language
url https://arxiv.org/abs/2604.12162