Explorerโ€บData Scienceโ€บMachine Learning
Research PaperResearchia:202610.12094

Returning to Development States: When Full Sandbox Snapshots Help Coding Agents

Daniel Thi Graviet

Abstract

Coding agents often make useful progress before they fail, but retry-from-scratch discards the machine state created along the way. This makes failure recovery an interesting search problem: if intermediate environment states preserve meaningful progress, restoring and branching from them could outperform repeatedly restarting from the initial state. We implement Go-Explore-style search over full Daytona snapshots and find that higher and more efficient solve rates depend mainly on progress that...

Submitted: October 12, 2026Subjects: Machine Learning; Data Science

Description / Details

Coding agents often make useful progress before they fail, but retry-from-scratch discards the machine state created along the way. This makes failure recovery an interesting search problem: if intermediate environment states preserve meaningful progress, restoring and branching from them could outperform repeatedly restarting from the initial state. We implement Go-Explore-style search over full Daytona snapshots and find that higher and more efficient solve rates depend mainly on progress that sits outside the repository, such as databases, services, and other runtime state. Our results suggest that full-snapshot search is most effective when long-horizon coding work spans multiple services whose state cannot be reconstructed from a single diff.


Source: arXiv:2610.13841v1 - http://arxiv.org/abs/2610.13841v1 PDF: https://arxiv.org/pdf/2610.13841v1 Original Link: http://arxiv.org/abs/2610.13841v1

Please sign in to join the discussion.

No comments yet. Be the first to share your thoughts!

Access Paper
View Source PDF
Submission Info
Date:
Oct 12, 2026
Topic:
Data Science
Area:
Machine Learning
Comments:
0
Bookmark
Returning to Development States: When Full Sandbox Snapshots Help Coding Agents | Researchia