Bruno Pedro's public notes

Can AI be the spark of non-technical API Design?

In 2019 I wrote a piece where I asked why weren’t companies building API Design tools that make it easier for non-technical people to collaborate in API Design activities. My hunch was that there wasn’t a market big enough to justify the investment in those tools.

Fast forward to 2023 where, according to recent surveys, about 50% of all API consumers aren’t engineers, and around 20% of decision-makers in large companies believe APIs are an integral part of their product. It’s clear that things are changing and APIs are no longer something exclusively technical.

Enter AI, particularly ChatGPT, and you see how it’s possible for a non-technical audience to engage in API Design even without knowing anything about OpenAPI or any other specification. About a year ago, when ChatGPT was still in its infancy, I did a demo video showing how easy it is for anyone to create an API from scratch.

However, ChatGPT is not the ideal interface for engaging in API Design activities. That is why I’m so excited to see API companies paying special attention to offering dedicated products that enable non-technical folks.

I just came out of a call with Aidan Cunniffe from Optic where we talked about their latest offering: LintGPT. It’s a tool that lets anyone write in plain English the linting rules they want to apply to their APIs. Their system then uses AI to “understand” those rules and apply them to existing API definitions.

This is a big deal because suddenly non-technical people, e.g. a product manager, can easily collaborate in the Design of an API. When collaboration around API Design permeates different functions in a company the result will be more consumer-centric and the business results will certainly improve.

I’ll be writing more about LintGPT and this trend I’ve been seeing so stay tuned.