Hardening Execution Sandboxes for Autonomous Coding Agents
How to safely let LLMs execute bash, edit files, and launch dev servers without risking host compromise, infinite network loops, or accidental data loss.
02/ Technical Writing & Research
Notes from the workbench.
First-hand engineering write-ups on building AI-native development workspaces, runtime sandbox isolation, autonomous state recovery, and empirical model benchmarks.
CraftCode Architecture: Building a Deterministic AI-Native Development Harness
Lessons from building CraftCode's multi-agent execution harness: deterministic tool orchestration, sandboxed runtime isolation, state recovery, and avoiding brittle autonomous loops.
Benchmarking Sub-30B Coding Models on Real-World Refactoring Tasks
A rigorous evaluation of small open-weight coding models on multi-file code navigation, AST manipulation, and edge-case test generation.