VeRO: An Evaluation Harness for Agents to Optimize Agents
This work addresses a gap in evaluating coding agents for iterative improvement, which is incremental but important for researchers in AI and agent development.
The paper tackles the lack of systematic evaluation for coding agents in agent optimization tasks, introducing VeRO as an evaluation harness and benchmark suite, and conducts an empirical study showing which modifications improve agent performance.
An important emerging application of coding agents is agent optimization: the iterative improvement of a target agent through edit-execute-evaluate cycles. Despite its relevance, the community lacks a systematic understanding of coding agent performance on this task. Agent optimization differs fundamentally from conventional software engineering: the target agent interleaves deterministic code with stochastic LLM completions, requiring structured capture of both intermediate reasoning and downstream execution outcomes. To address these challenges, we introduce VERO (Versioning, Rewards, and Observations), which provides (1) a reproducible evaluation harness with versioned agent snapshots, budget-controlled evaluation, and structured execution traces, and (2) a benchmark suite of target agents and tasks with reference evaluation procedures. Using VERO, we conduct an empirical study comparing optimizer configurations across tasks and analyzing which modifications reliably improve target agent performance. We release VERO to support research on agent optimization as a core capability for coding agents.