activity
20222026
most citedAll You Need Is Logs: Improving Code Completion by Learning from Anonymous IDE Usage Logs

1 citations · 1 across the 4 of their papers we have counts for

collaborators
Showing cs.AIShow all

1 paper · 1 filter

cs.AI2026

AgentLens: Production-Assessed Trajectory Reviews for Coding Agent Evaluation

Andrey Podivilov, Vadim Lomshakov, Sergey Savin +4

We present AgentLens, a production-assessed benchmark for interactive code agents. Most code-agent benchmarks reduce a run to a single bit -- did the task pass? -- but the people w…