# Taming Uncertainty via Automation: Observing, Analyzing, and Optimizing Agentic AI Systems

arXiv: https://arxiv.org/abs/2507.11277 · HTML: https://mlsanity.com/paper/2507.11277/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 8 of its 29 references that are in the map.
- First posted on arXiv 15 Jul 2025.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 8 of 8)

- [Model Context Protocol (MCP): Landscape, Security Threats, and Future Research Directions](https://mlsanity.com/paper/2503.23278/) (2025)
- [TRAIL: Trace Reasoning and Agentic Issue Localization](https://mlsanity.com/paper/2505.08638/) (2025)
- [Non-Determinism of "Deterministic" LLM Settings](https://mlsanity.com/paper/2408.04667/) (2024)
- [ITBench: Evaluating AI Agents across Diverse Real-World IT Automation Tasks](https://mlsanity.com/paper/2502.05352/) (2025)
- [Beyond Black-Box Benchmarking: Observability, Analytics, and Optimization of Agentic Systems](https://mlsanity.com/paper/2503.06745/) (2025)
- [Agentic AI Process Observability: Discovering Behavioral Variability](https://mlsanity.com/paper/2505.20127/) (2025)
- [The WHY in Business Processes: Discovery of Causal Execution Dependencies](https://mlsanity.com/paper/2310.14975/) (2023)
- [Detecting Anomalous Events in Object-centric Business Processes via Graph Neural Networks](https://mlsanity.com/paper/2403.00775/) (2024)

## Built on by (most cited first, 12 of 12)

- [AgentSight: System-Level Observability for AI Agents Using eBPF](https://mlsanity.com/paper/2508.02736/) (2025)
- [Silenced Biases: The Dark Side LLMs Learned to Refuse](https://mlsanity.com/paper/2511.03369/) (2025)
- [Beyond Task Completion: An Assessment Framework for Evaluating Agentic AI Systems](https://mlsanity.com/paper/2512.12791/) (2025)
- [Security in the Age of AI Teammates: An Empirical Study of Agentic Pull Requests on GitHub](https://mlsanity.com/paper/2601.00477/) (2026)
- [AgentFixer: From Failure Detection to Fix Recommendations in LLM Agentic Systems](https://mlsanity.com/paper/2603.29848/) (2026)
- [Efficient Serving for Dynamic Agent Workflows with Prediction-based KV-Cache Management](https://mlsanity.com/paper/2605.06472/) (2026)
- [Transduction is All You Need for Structured Data Workflows](https://mlsanity.com/paper/2508.15610/) (2025)
- [Empowering Real-World: A Survey on the Technology, Practice, and Evaluation of LLM-driven Industry Agents](https://mlsanity.com/paper/2510.17491/) (2025)
- [VIGIL: Towards Edge-Extended Agentic AI for Enterprise IT Support](https://mlsanity.com/paper/2603.16110/) (2026)
- [Semantic Feature Analysis: Improving Agents Without Searching Over Rollouts](https://mlsanity.com/paper/2604.10513/) (2026)
- [Agent System Operations: Categorization, Challenges, and Future Directions](https://mlsanity.com/paper/2606.01581/) (2026)
- [AUDITA: certified auditing and causal attribution of adverse outcomes in autonomous multi-agent systems](https://mlsanity.com/paper/2608.22160/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) are increasingly deployed within agentic systems - collections of interacting, LLM-powered agents that execute complex, adaptive workflows using memory, tools, and dynamic planning. While enabling powerful new capabilities, these systems also introduce unique forms of uncertainty stemming from probabilistic reasoning, evolving memory states, and fluid execution paths. Traditional software observability and operations practices fall short in addressing these challenges. This paper presents our vision of AgentOps: a comprehensive framework for observing, analyzing, optimizing, and automating operation of agentic AI systems. We identify distinct needs across four key roles - developers, testers, site reliability engineers (SREs), and business users - each of whom engages with the system at different points in its lifecycle. We present the AgentOps Automation Pipeline, a six-stage process encompassing behavior observation, metric collection, issue detection, root cause analysis, optimized recommendations, and runtime automation. Throughout, we emphasize the critical role of automation in managing uncertainty and enabling self-improving AI systems - not by eliminating uncertainty, but by taming it to ensure safe, adaptive, and effective operation.