A detailed analysis of agentic coding in legacy codebases documents several migration efforts and their outcomes1. Teleport spent a quarter with 13 engineers pointing frontier models at its codebase, building a multi-agent harness, yet a plain prompt beat it in finding a critical vulnerability. Shopify rebuilt its Shop consumer app from React Native to native Swift and Kotlin in twelve weeks using agent-gated checkpoints. Asana cleared a multi-year Enzyme backlog in two weeks for roughly $12,000 in model and infrastructure cost. Across 520 agent runs on SWE Refactor Bench, only 28 passed migration audit, behavioral tests, and independent verification.
Brownfield Agentic Engineering Guide Details Migration Pitfalls, Case Studies
Analysis of agentic coding in legacy codebases covers Teleport, Shopify, and Asana migration efforts, plus SWE Refactor Bench results showing 28 of 520…