DOP 343: Your APIs Were Never Built to Be the Front Door

Episode 343

Show Notes

#343: Here’s the thing about your company’s APIs – they were built for your own engineers to use inside your own software. Nobody designed them to be the front door. But that’s exactly what’s happening. Matt DeBergalis, CEO of Apollo GraphQL, makes a pretty compelling case that AI agents are turning internal APIs into the actual interface between companies and customers. Not the website. The APIs themselves.

And most of them aren’t ready for that. At all.

Think about what happens when you point a model at a typical REST API. GitHub’s API returns hundreds of fields for a single repository object. Fine when another service is calling it. But a model? All those extra fields are context you’re paying for, and they make the model hallucinate. Matt says you need something between the model and all those backend services – an orchestration layer that takes one request and handles the mess underneath. That’s where GraphQL comes in.

He draws a parallel that’ll land immediately if you’ve been in this space a while. APIs right now are pets – handwritten, named, carefully managed. But AI-generated code is about to produce way more microservices, which means way more APIs. They’re going to become cattle. And just like containers needed Kubernetes, APIs are going to need declarative infrastructure to manage them at scale.

The conversation takes an interesting turn when Darin pushes back on the idea that developers are becoming architects. His take: we’re becoming product managers. Matt says both. Viktor throws in code reviewers. Matt’s own story backs it up – he codes more as CEO than he did as CTO, because AI handles the parts he never had time to learn. He doesn’t know modern React. Doesn’t need to.

One more thing that should make any tech company uncomfortable: if AI agents are how customers find you now, what happens to your docs-page-driven acquisition funnel? Apollo’s already made the shift – their first audience for documentation is the models, not the humans.

Frequently Asked Questions

Are REST APIs the right interface for AI agents to use?

Matt DeBergalis, CEO of Apollo GraphQL, argues on DevOps Paradox episode 343 that they were never built for it. REST APIs were designed for trusted engineers calling them as part of an internal implementation, where latency matters little and the caller discards whatever fields it does not need. He points at GitHub’s REST API returning hundreds of fields for one repository object, all of which becomes context a model pays for.

Why does GraphQL suit AI agents better than REST?

Matt DeBergalis explains on DevOps Paradox episode 343 that GraphQL distils complexity into a short query and pushes the rest into infrastructure, the way a database engine hides a query planner. A model writes ten or twenty lines rather than a thousand lines of procedural code. It also avoids making the model orchestrate the three, four, or ten separate calls a resource-shaped REST API would otherwise require.

Should AI agents call APIs non-deterministically?

Matt DeBergalis argues on DevOps Paradox episode 343 that you want the non-determinism at the experience layer and precision underneath it. His example is a bank, where every customer asking about their account should get the same five recent transactions rather than sometimes ten, with no freelancing about how systems get combined. Viktor Farcic pushes back, asking why AI needs to be involved in the deterministic parts at all.

Will AI change how many APIs a company has?

Matt DeBergalis predicts on DevOps Paradox episode 343 that companies are about to own many more microservices and APIs, because asking an agent to write code works best when that code is structured as independent modules. He draws the parallel to containers, noting Kubernetes appeared because teams had two thousand of them rather than two, and expects APIs to become similarly ephemeral and to need orchestrating infrastructure.

What is agent experience and how does it differ from developer experience?

Matt DeBergalis describes it on DevOps Paradox episode 343 as writing documentation and building products with the models as the first audience rather than people. At Apollo GraphQL that has meant making sure agents can interact with the product, and reworking how GraphQL reports errors on the assumption an agent rather than a human is reading the output. Humans still use it, so both audiences have to be served together.

What happens to companies whose documentation is the top of their funnel?

Matt DeBergalis is blunt about this on DevOps Paradox episode 343: if your business model depends on human developers visiting your docs page to work something out, you have a problem. He compares the shift to what SEO once meant for many kinds of company, and argues it now cuts deeper, because whether a customer finds you at all depends on what a model says while they type.

What is the DevOps Paradox podcast?

DevOps Paradox is a weekly podcast co-hosted by Darin Pope and Viktor Farcic, covering DevOps, platform engineering, and modern software delivery. Episode 343, “Your APIs Were Never Built to Be the Front Door,” features Matt DeBergalis of Apollo GraphQL on why APIs designed for internal engineers do not survive contact with AI agents. Every episode page carries audio, video, and a full transcript.

Episode Transcript

Read the Full Transcript

Topics

Share and Download

Guests

Matt DeBergalis

Matt DeBergalis

Matt DeBergalis is the Chief Executive Officer and Co-Founder of Apollo GraphQL, focused on bringing the popular GraphQL technology to the enterprise. He previously served as Apollo’s CTO, leading product and engineering. Matt’s longtime focus has been in open source and platforms: he co-founded Meteor.js, which grew to become one of the most popular open-source projects in the world for developing full-stack web apps with JavaScript, as well as ActBlue, the American political fundraising platform that revolutionized grassroots political giving. He attended the Massachusetts Institute of Technology and is based in the San Francisco Bay Area with his family. In his spare time, Matt enjoys taking to the air and flying his 1966 Beechcraft Baron.

Hosts

Darin Pope

Darin Pope

Viktor Farcic

Viktor Farcic

Viktor Farcic is a member of the Google Developer Experts and Docker Captains groups, and published author.

His big passions are DevOps, Containers, Kubernetes, Microservices, Continuous Integration, Delivery and Deployment (CI/CD) and Test-Driven Development (TDD).

He often speaks at community gatherings and conferences.

He has published DevOps Paradox and Test-Driven Java Development.

His random thoughts and tutorials can be found in his blog The DevOps Toolkit.