Martin Fowler: DSLs provide a strong harness for reliable LLM code generation

DSLs Enable Reliable Use of LLMs

Unmesh Joshi argues on Martin Fowler's blog that upfront specs are just hypotheses—design is discovered through implementation. DSLs and domain abstractions act as a harness for LLMs, setting clear boundaries so generated code matches intent. The Tickloom example shows a domain model for distributed systems where an LLM helps iteratively build the DSL and then serves as a natural-language interface to it. The DSL becomes the source of truth for the system.

Why it matters: A solid engineering pattern piece on Martin Fowler's blog—using DSLs to bound LLM code generation is a practical take with a worked example. Capped at 72 because it's an opinion/pattern article, not a product or model release, and the audience fit skews toward backend architects.