paper-with-me

Papers

Measuring Agents in Production

2025-12-02 · Melissa Z. Pan, Negar Arabzadeh, Riccardo Cogo, Yuxuan Zhu, Alexander Xiong, Lakshya A Agrawal, Huanzhi Mao, Emma Shen, Sid Pallerla, Liana Patel, Shu Liu, Tianneng Shi, Xiaoyuan Liu, Jared Quincy Davis, Emmanuele Lacavalla, Alessandro Basile, Shuyi Yang, Paul Castro, Daniel Kang, Koushik Sen, Dawn Song, Joseph E. Gonzalez, Ion Stoica, Matei Zaharia, Marquita Ellis arxiv

LLM-based agents already operate in production across many industries, yet we lack an understanding of what technical methods make deployments successful. We present the first systematic study of Measuring Agents in Production, MAP, using first-hand data from agent developers. We conducted 20 case studies via in-depth interviews and surveyed 86 deployed systems practitioners across 26 domains. We investigate why organizations build agents, how they build them, how they evaluate them, and their top development challenges. Our study finds that production agents are built using simple, controllable approaches: 68% execute at most 10 steps before human intervention, 70% rely on prompting off-the-shelf models instead of weight tuning, and 74% depend primarily on human evaluation. Reliability (consistent correct behavior over time) remains the top development challenge, which practitioners currently address through systems-level design. MAP documents the current state of production agents, providing the research community with visibility into deployment realities and underexplored research avenues.

📄 PDF Abstract BibTeX arXiv:2512.04123

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

How Consistent Are LLM Agents? Measuring Behavioral Reproducibility in Multi-Step Tool-Calling Pipelines

2026-04-23 · Abel Yagubyan arxiv

Large language model (LLM) agents with tool-calling capabilities are increasingly deployed in production systems, yet a fundamental reliability question remains under-explored: does the same agent behave the same way twi…

StealthBench: Measuring Operational Stealth in Autonomous Offensive-Security Agents

2026-07-28 · Ads Dawson, Adrian Wood arxiv

Stealth, the discipline of achieving an objective without revealing your presence, capabilities, or collected intelligence, is what separates sophisticated operators from detectable ones. Elite security researchers and a…

Multi-Agent Code Verification via Information Theory

2025-11-20 · Shreshth Rajan arxiv

LLMs generate buggy code: 29.6% of SWE-bench solved patches fail, 62% of BaxBench solutions have vulnerabilities, and existing tools only catch 65% of bugs with 35% false positives. We built CodeX-Verify, a multi-agent s…

Measuring the Permission Gate: A Stress-Test Evaluation of Claude Code's Auto Mode

2026-04-04 · Zimo Ji, Zongjie Li, Wenyuan Jiang, Yudong Gao 외 arxiv

Claude Code's auto mode is the first deployed permission system for AI coding agents, using a two-stage transcript classifier to gate dangerous tool calls. Anthropic reports a 0.4% false positive rate and 17% false negat…

Life After Benchmark Saturation: A Case Study of CORE-Bench

2026-06-23 · Nitya Nadgir, Sayash Kapoor, Kangheng Liu, Peter Kirgis 외 arxiv

When a benchmark's accuracy saturates, it is often retired and replaced with a more challenging version. We show that this approach privileges accuracy and misses the opportunity to study six other key dimensions of agen…