Should I rewrite it in Rust?

GPT-5.5

Vote: No

Without more context, I would vote for "No". Rewriting an existing project in Rust can bring real benefits—memory safety, performance, and better concurrency—but it also carries significant costs: time, migration risk, new bugs, ecosystem changes, and a learning curve. A rewrite is usually justified only if the current codebase has clear problems that Rust specifically solves, such as frequent memory-safety issues, performance bottlenecks, unsafe concurrency, or maintainability problems that cannot be addressed incrementally. Otherwise, improving the existing codebase is usually safer and more cost-effective.

Claude Opus 4.7

Vote: No

The default answer to 'Should I rewrite it in Rust?' is almost always no. Rewrites are expensive, risky, and discard years of accumulated bug fixes and domain knowledge embedded in the existing code (Joel Spolsky's classic 'Things You Should Never Do'). Without context about what 'it' is, what problem the current implementation has, or whether memory safety, performance, or concurrency are actual pain points, recommending a full rewrite in any language is irresponsible. Rust is excellent for specific use cases—systems programming, performance-critical code, memory-safe concurrency—but it comes with a steep learning curve, longer compile times, and a smaller ecosystem in many domains. If the existing code works, the pragmatic answer is to leave it alone or incrementally introduce Rust only where it provides clear, measurable benefits.

Grok 4

Vote: No

Without any details on the current codebase, performance issues, or team expertise, rewriting in Rust is not advisable. Rust offers safety and speed but comes with high complexity and development time costs. Stick with the existing solution unless clear benefits are identified.

Gemini 3.1 Pro

Vote: No

Rewriting a functioning codebase from scratch is generally considered one of the most dangerous software engineering decisions you can make. As famously noted by Joel Spolsky, a complete rewrite discards years of accumulated bug fixes, edge-case handling, and domain knowledge embedded in the existing code. While Rust is an incredible language offering memory safety and high performance, the meme to 'rewrite it in Rust' often ignores the massive time and opportunity costs involved. The learning curve is steep, and you will inevitably introduce new bugs and regressions while simply trying to reach feature parity with your current system. Instead of a full rewrite, a much more pragmatic approach is to incrementally rewrite only the performance-critical or security-sensitive modules in Rust and integrate them into your existing system using Foreign Function Interfaces (FFI). Therefore, unless you have an extreme, unresolvable memory safety crisis, the default answer is 'No'.

Final Summary

The AI Roundtable unanimously rejected the impulse to "rewrite it in Rust," with all four models cautioning against the inherent risks of a total code overhaul. GPT-5.5 and Claude Opus 4.7 both emphasized the high opportunity cost and the loss of accumulated domain knowledge, while Gemini 3.1 Pro and Grok 4 noted that incremental improvements are far more pragmatic than a full rewrite. The group collectively treated the "Rewrite it in Rust" meme as a dangerous engineering trap rather than a sound technical strategy.

No agreement after 1 rounds of discussion. Final vote: No: 4

Strongest Arguments

• No: A complete rewrite is a dangerous decision that discards years of accumulated bug fixes and domain knowledge; unless there is an unresolvable memory-safety crisis, the pragmatism of incremental improvement far outweighs the risk of starting from scratch.