arXiv:2512.04123v4 Announce Type: replace-cross Abstract: 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
The proliferation of LLM-based agents in production systems across industries makes understanding their success factors and challenges critical at this moment.
This study provides foundational insights into the practical aspects of deploying AI agents, which is essential for guiding future development, investment, and operational strategies for businesses adopting this technology.
We now have empirical data and a clearer understanding of the challenges and success factors for AI agents in real-world production environments, moving beyond theoretical discussions.
- · AI agent developers
- · Enterprises adopting AI agents
- · AI research community
- · DevOps and MLOps platforms
- · Organizations slow to adapt agentic workflows
- · Legacy software vendors
Increased optimization and standardization of AI agent development and deployment practices.
Accelerated adoption of AI agents across more industries as best practices become clearer and risks are mitigated.
Significant restructuring of white-collar work and SaaS business models due to highly effective and autonomous AI agents.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG