Confidence: high (within bounds). Evidence: case study and benchmark. Last substantive change: 2026-07.
This subsystem owns the act of building: writing new code, changing existing code, reviewing it, and converting one codebase into another. It is the part of the factory with the most public proof, and the clearest limits.
The conclusion
Agents are already capable of substantial greenfield implementation, maintenance, review, and codebase conversion when the work has strong oracles and controllable seams. They are weaker on ambiguous product behavior and safety-critical novelty, where there is no oracle to check against.
How the thinking got here
Bounded patch benchmarks gave way to multi-file features, then to million-line greenfield products, then to large-scale review and migration systems. The largest published migrations rewrote entire runtimes with a conformance suite as the oracle, which is exactly the condition that makes agent implementation trustworthy: a language-independent check the agent cannot game.
Credible alternatives, and when each is right
| Approach | Right when |
|---|---|
| Human implementation with AI assistance | novel or safety-critical work |
| Agent implementation with human review | moderate risk, reviewable diffs |
| Agent implementer plus independent agent reviewer | scale with context-separated checks |
| Regeneration rather than migration | the spec is clearer than the legacy code |
| Mixed criticality lanes | different risk tiers in one system |
Where it fails and what we still don't know
The strongest results come from greenfield scaffolding, bounded changes, review, and conversions with mature test suites. The weakest come from ambiguous product behavior and safety-critical novelty. Open questions include net-new complex systems, hardware and embedded work, performance engineering, data-heavy systems, and maintenance beyond the first year.
What would change our mind
Sustained evidence of agents building novel, safety-critical systems without a pre-existing oracle would extend the boundary well past where it sits today.