Aug 2026· Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2· pp. 13289-13290· 0 citations· 1 references
Abstract
As Large Language Model (LLM) agents transition from general-purpose assistants to specialized enterprise tools, grounding them in industrial reality remains a significant challenge. To address this, we developed AssetOpsBench, a comprehensive framework for the lifecycle of AI agents in Industrial Asset Operations and Maintenance, built around the Model Context Protocol (MCP) as the standard for connecting agents to complex enterprise data silos. In this hands-on tutorial, participants will learn to build reliable industrial agents using MCP and AssetOpsBench. The session is divided into two parts. In the first part, we dive into an end to end MCP grounded agent pipeline that connects specialized MCP servers to high velocity industrial data including time series telemetry, IoT streams, failure mode records, and work order histories while orchestrating workflows with Plan Execute and Reflexion planners. In the second part, we unlock predictive analytics such as anomaly detection and Remaining Useful Life (RUL) prediction within agentic workflows, assess agent reliability through multi-dimensional evaluation, and ground agents against physical constraints. Whether for researchers or practitioners, this tutorial provides the foundations for building auditable, production-grade agents for Industry 4.0. AssetOpsBench is accessible at: https://github.com/IBM/AssetOpsBench.
Results suggest that localized multi-agent automation can make time-series modeling easier, and significantly reduces measured per-run cost, while staying within a comparable energy consumption.
Large language models (LLMs) have evolved from standalone generative systems into agentic AI systems capable of planning, reasoning, tool use, and multi-agent collaboration. Enterprises are increasingly adopting AI agents to automate and orchestrate complex workflows, from IT operations to employee productivity. While early deployments focused on proof-of-concept prototypes, the past year has marked a clear shift toward production-grade enterprise AI agents. This transition has been enabled by a wave of new technologies, including multi-agent orchestration, memory and state management, skill-based and modular agent architectures, and deeper integration with enterprise data and workflow platforms, which together make scalable, reliable agent systems feasible in practice. At the same time, moving agents into production introduces new technical and organizational challenges, such as rigorous evaluation and benchmarking, security and governance, and system design for long-running, autonomous operation. Building on the success of our two prior highly attended editions: ''Agentic AI for Enterprise'' workshop at KDD 2025 and ''Enterprise RAG'' workshop at CIKM 2024, this workshop aims to bring together researchers and practitioners to examine how enterprise AI agents can successfully move from prototypes to production. We focus on three pillars: 1) Agent architectures and systems; 2) Enterprise applications and deployments; 3) Evaluation and governance.
Min Du, Anbang Xu, Jasmine Jaksic et al.· Proceedings of the 32nd ACM...· 0 citations
This work presents ComponentBench, a benchmark and diagnostic pipeline for component-level evaluation of computer-use agents on modern web UIs, and introduces a scalable pipeline for auditing realized structural difficulty after implementation and synthesizing structured failure analyses across tasks and component families.
Tianchen Guan, Xinlei Lin, Royce Cheng-Yue et al.· 0 citations
Enterprises have no shortage of agentic AI demonstrations; what they lack is a repeatable way to turn a demonstration into a system a business can depend on. Forward Deployed Engineering closes that gap by embedding an engineer with the customer to own the path from prototype to hardened, monitored production. This paper presents the complete Forward Deployed Engineering (FDE) pipeline that the AlgoProfessor team uses to solve modern enterprise agentic AI challenges, and makes its reliability core precise and reproducible. The central obstacle is compounding error: a task of k sequential steps, each succeeding with probability p, succeeds end-to-end with probability p to the power k, which collapses as k grows. We give a taxonomy of enterprise challenges, a staged pipeline closed by a continuous improvement loop, and a reference architecture for a hardened agentic system in which every action passes verification and human oversight and observability cross-cut the whole. We prove two guarantees: verification with retry lifts the effective per-step success from p to p over one minus one minus p times the catch rate, moving the base of the exponential, and a fixed verification budget is best spent on the weakest steps. A fully reproducible simulation confirms the model: without verification, end-to-end success decays as p to the power k, collapsing to 0.12 at twenty steps, while per-step verification holds it at 0.64 and checkpoints at 0.94; greedy allocation to the weakest steps reaches 0.65 at a budget where random placement reaches only 0.31; and the improvement loop raises reliability from 0.22 toward target over successive iterations. Dependable enterprise agentic AI is engineered, not prompted, and the FDE pipeline is how that engineering is organised.
S. Satyanarayana· International Journal of Com...· 0 citations
DataClawEval is introduced, the first comprehensive benchmark designed specifically to evaluate the end-to-end task completion capabilities of autonomous agents in real-world data engineering scenarios, and it comprises 100 rigorous, end-to-end tasks spanning five execution engines.
The more typical feature of agentic AI systems is dynamic, multistep workflows where autonomous components plan, reason, and communicate with external tools and data sources in a series of iterations. Such flexibility increases capability but also brings nondeterminism which is inherent and where the same inputs can result in different execution paths and outputs. The variability creates a major challenge to the traditional observability approaches that are mostly created to support deterministic and service-oriented architectures. This paper redefines observability as an architectural element and introduces a trace-oriented architecture to suit agentic processes. The suggested solution presents semantically rich trace units capturing reasoning transitions, the intent to select a tool, the evolution of memory, and policy interactions, to gain a better insight into the execution behaviour. One of the major innovations is the combination of branch-aware trace modelling with an adaptive fidelity mechanism to dynamically change monitoring granularity based on uncertainty and anomaly indicators. Experimental analysis of various agent workflows shows significant increases in completeness of traces, accuracy in detection of anomalies and localization of root-cause, and a significant decrease in diagnosis time. The findings suggest that the suggested architecture does not only increase interpretability but also helps to ensure the reliable and efficient functioning of non-deterministic AI systems.
Ankur Gupta, Karan Gupta, Divyakumar Deepak Savla et al.· International Conference on...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
