PlanLayer: Making the Plan the Unit of Accounting for Agentic AI [Short]
Hannah B. Pasandi, Negar Arabzadeh, Melissa Pan (UC Berkeley)
[Video]

Learning on the GPUs the World Owns: A Fleet Contract for Learned Systems on Commodity Hardware [Position Paper]
Hannah B. Pasandi (UC Berkeley), Mohammad Hosseini (Shahid Beheshti University), Sina Darabi (Università della Svizzera italiana)
[Video]

The Monolingual Assumption: What Web Agents Pay to Work Beyond English [Short]
Hannah B. Pasandi (UC Berkeley), Haniyeh B. Pasandi (University of Maryland, College Park)
[Video]

Defining AI-Native Systems: Autonomy as the Authority to Revise [Position Paper]
Cheng Tan (Northeastern University)
[Video]

Understanding the Use and Impact of AI-Coding Assistants and Agents in System Development
Weidong Zhu, Arjun Arunasalam, Chen Chen, Raju Rangaswami (Florida International University)
[Video]

AgentLoupe: Kernel-Resident Semantic Causal Labels for AI Agent Observability
Yifei Bao, Ting Liu (Southeast University), Yuxin Ren, Ruizhe Huang (Peking University), Likang Luo, Ziteng Xu, Yu Wang (Ant Group), Peng Jiang (Southeast University)
[Video]

Crash-Safe Effects for Agents in Operational Systems [Position Paper]
Hannah B. Pasandi (UC Berkeley), Mohammad Sepahi (Old Dominion University)
[Video]

Agents Break the Scheduler: Why ML-Based Resource Managers Fail for Agentic Workloads
Sheetal Sriram (Georgia Institute of Technology)
[Video]

Using Epistemic Observability in Agentic Systems for Combating Hallucinations
Tony Mason (University of British Columbia), Vaastav Anand (Max Planck Institute for Software Systems)

Don’t Trust the Code, Check Its Effects: Runtime Refinement for Regenerated Systems Code Under an Adversarial Generator
Jinhao Hu (Max Planck Institute for Software Systems), Ashvin Goel (University of Toronto), Laurent Bindschaedler (Max Planck Institute for Software Systems)

Survival of the Luckiest: The Impacts of Noisy Evaluation in Agentic Evolution for Systems
Johannes Freischuetz (University of Wisconsin–Madison), Venkatesh Emani (Microsoft – Gray Systems Lab), Shivaram Venkataraman (University of Wisconsin–Madison)

Argo: Efficient Importance Labeling for Enterprise Email Systems
Siddhant Ray (University of Chicago), Ganesh Ananthanarayanan, Kevin Chian, Yan Guo, Cristina St Hill (Microsoft), Jack W. Stokes (Independent), Victor Wang (Microsoft), Junchen Jiang (University of Chicago / Tensormesh)