Keynote | Agents Need Management Too

Keynote | Agents Need Management Too

by Julian Duru

Applied Sciences & Engineering AI — News & Applications UYComputer scienceUYQArtificial intelligence (AI)
🎙 Julian Duru 👥 278 📅 August 8, 2026 ⏱ 38 min 👁 6 📄 expert opinion 🧭 2026-08-16
Available in: English (current) Français

Keywords

AI agentsagent managementverifiable workdecentralized systemssmall business

Summary

Julian Duru, co-founder of Flow Research, delivers a keynote on the necessity of managing AI agents in the context of human-agent collaboration. He draws parallels between managing human employees and managing agents, emphasizing the need for verification, accountability, and evidence in agent actions. Duru discusses the architecture of agent systems, the importance of understanding trade-offs, and the potential for open markets where agents trade services. He highlights the benefits for small businesses, which could leverage specialized agents without high costs. The talk touches on digital autonomy and the economics of useful work, but remains largely conceptual. Duru’s background at Moniepoint and his focus on decentralized systems inform his perspective. The presentation is informal, with audience interaction, but lacks concrete examples or technical depth.

123 words

Critical Evaluation

Value of the Information & Strength of the Argument

The talk offers a high-level vision of agent management, drawing analogies to human resource management. The argumentation is based on the speaker’s experience and intuition rather than empirical evidence. Key concepts like verifiable work and accountability are introduced but not deeply explored. The value lies in framing the problem of agent governance, but the argumentation is weak due to lack of concrete examples or case studies.

Scientific Rigor, Source Quality, Title Accuracy

The talk has no citations or references, and the description provides no links. The speaker’s credibility is established through his background, but the content lacks scientific rigor. The title is appropriate, but the content is more of a vision statement than a detailed analysis. No comments were provided for analysis.

131 words

Title / Content Match

The title accurately reflects the core theme of managing AI agents, though the talk is more of a high-level vision than a detailed management framework.

Quality & Reliability

5/10

The talk is an expert opinion with limited technical depth, no citations, and a fragmented delivery. The speaker's background adds credibility, but the content lacks rigor and verifiable sources.

Key Moments

Markers derived by PSI from the transcript: the creator did not define chapters.

Contribution & Novelties

The talk contributes a perspective on managing AI agents as a critical component of AI deployment, emphasizing verification and accountability. It introduces the concept of open agent markets and the potential for small businesses to leverage AI. However, the ideas are not novel and lack depth.

Pour aller plus loin :

79 words

Radar Profile

The radar profile shows low scores across all dimensions, indicating a talk with limited information density, technical depth, and reliability. The speaker's vision is present but not substantiated.

Reliability 4/10