Building developer community and trust in today’s AI-first world
AI can scale how we support developers, but it can’t be the support.
A guest article by Sanjay Sarathy, VP Developer Experience and Self-Service at Cloudinary.
Not long ago, developer discovery was a pretty predictable process. You optimized for SEO, published quality content on YouTube, maybe answered questions on Reddit. While those channels remain crucial, something new is emerging.
At Cloudinary, we’ve noticed a subtle but meaningful shift in how developers find us. Until recently, around 65-70% of new developers discovered Cloudinary through search and video content. Then, earlier this year, we began to see more developers arriving via ChatGPT and other AI assistants. We’re not alone. AI tools and large language models (LLMs) are changing the discovery game for everyone.
This evolution is one that every developer experience leader now faces. AI copilots and LLMs are increasingly acting as both discovery engines and engagement environments, reshaping how developers encounter, learn, and build with your product.
AI assistants are becoming primary discovery channels for developers
For years, developer discovery followed a simple pattern: search, click, learn, build. That flow now also begins with a question in an AI chat window.
Developers aren’t abandoning search engines; they’re simply adding new tools to the mix. Alongside Google results and YouTube tutorials, copilots and chat assistants are becoming powerful new ways to surface snippets, integrations, and documentation.
This “AI-assisted front door” means that content doesn’t just need to rank well. It needs to resonate with both humans and machines. The quality, structure, and credibility of technical content now influence whether AI tools can surface and trust it.
We began seeing evidence of that shift in our own data: Well-structured, developer-trusted documentation was being surfaced in AI-generated responses. That trend didn’t come from deliberate optimization, but from the strength and clarity of the content itself. It’s a reminder that credibility continues to matter, regardless of the discovery channel.
Docs are the front door for AI tools
Documentation has always been the heartbeat of developer experience. In the AI era, it’s also becoming a gateway through which many developers first encounter a product.
To support both traditional SEO and AI ecosystems, we’ve been modernizing our docs infrastructure in key ways:
Dual-structured content. We now maintain parallel Markdown and HTML structures. This ensures our content is machine-readable and easily parsed by AI agents while still serving human readers.
Reducing hallucinations. Developers are bullish on AI tools, but as Prashanth Chandrasekar, CEO of Stack Overflow, recently noted, many still hesitate to fully trust their outputs. To help close that gap, Cloudinary documentation is structured so it can serve as a source of truth for AI assistants, ensuring their guidance stays aligned with our official SDKs and APIs.
Mixed-mode readiness. We assume developers will move fluidly between traditional search, docs, and AI copilots, so we optimize for all three.
Treating documentation as a product (not just a reference) has helped us stay credible with both developers and the AI tools they increasingly depend on.
Developer community now spans human connections and AI-mediated experiences
If discovery is evolving, so is community. The definition of “developer community” can no longer be confined to message boards, Discord groups, and YouTube comments.
Developers now learn, build, and troubleshoot in conversation with copilots often without posting a question or reading a thread. That doesn’t mean human connection has disappeared; it means community now spans both human and AI-mediated experiences.
At Cloudinary, we still believe in cultivating human relationships, the trust, empathy, and collaboration that no model can replace. AI can scale how we support developers, but it can’t be the support. That belief drives us to listen directly to developers, not just through metrics but through conversations and outreach.
Earlier this year, we asked our community how we could make their experience with Cloudinary better, and more than 100,000 developers responded. It’s a reminder that when you put in the work to earn trust and demonstrate that feedback will lead to real improvements, developers want to share their insights. That kind of engagement can’t be automated; it has to be earned.
And that’s really the point. The strength of any developer ecosystem still depends on people. AI can help make those interactions more efficient and scalable, but it doesn’t replace the relationships that create them. While AI can recommend Cloudinary, recommendations from the actual community cements trust. AI simply helps scale the experiences that they create.
How to adapt your developer program for AI-assisted discovery
Adapting to this AI-assisted landscape doesn’t mean reinventing your developer program. It means evolving your metrics, mindset, and methods.
Here’s what forward-looking developer experience leaders can do today:
Redefine success. Track how often your docs or code samples appear in AI outputs, which are a new indicator of visibility and trust. Continue monitoring API usage, SDK adoption, and feature utilization.
Design for a mixed-mode world. Assume developers will move between human, self-serve, and AI-driven experiences, and make sure each is equally supported.
Use feedback data intentionally. Developers willingly share insights when they see it improves their experience.
Keep humans at the center. AI can assist and amplify, but developer success still depends on human empathy, clarity, and trust.
AI amplifies developer success without replacing human relationships
Developers are natural experimenters. They’re always looking for new, faster, smarter ways to build. AI represents the latest evolution of that spirit.
For developer experience leaders, the opportunity is clear: Use AI as a tailwind to scale great experiences, not as a replacement for the relationships that make them meaningful.
At Cloudinary, our north star hasn’t changed. Whether a developer finds us through SEO, YouTube, or ChatGPT, our goal is the same, to help them succeed. Because when developers succeed, everything else follows.
Solid framework for adapting devrel in the AI era. The shift from pure SEO to AI-assisted discovery is real, and treating docs as machine-readable first-class products is dunno the move most teams will miss. What's interesting is how this redefines community engagement metrics when developers troubleshoot silently with copilots instead of posting. The 100k feedback response shows trust still comes from human touchpoints not automation alone.