What LLMs and Biology Have in Common | Max Agency
Working with LLMs resembles experimental biology more than deterministic software engineering.
“They're poking a system they don't fully understand and watching how it responds.”
In a LangChain discussion, a speaker argues that understanding and using LLMs is more like biology research than traditional software engineering, since practitioners probe and perturb an opaque system rather than specifying deterministic behavior. The analogy extends to interpretability work at model labs and to large training runs whose outcomes are only observable after weeks or months. It's a useful framing for how engineers should approach AI systems, but not a major industry announcement.