
Semih Ozden
I have over six years of experience in designing and developing dynamic websites with interactive features that drive business growth and improve user experience. I use React.js with Next.js in frontend development, creating static site generation (SSG) and server-side rendering (SSR) pages, and refactoring projects from Angular.js to React.js. I also use Redux for state management, Cypress for end-to-end, integration, and component testing, and Tailwind.css and Material-UI for building custom themes.
I am passionate about web performance and SEO optimization, and I use Webpack, Gulp, Babel, and Eslint to implement best practices and solutions. I also work with API-centric design approach, use CSS3 and HTML5 for designing websites, and use JavaScript ES6+ features to develop the entire website. I have experience in checking pull requests, implementing Scrum strategies, and collaborating with other frontend teams.
The AI Call Is the Render: Server Components as the Intelligence Boundary
React Advanced 2026
Upcoming
The AI Call Is the Render: Server Components as the Intelligence Boundary

Most React applications treat AI as a separate service: a backend calls an LLM, returns JSON, and a client component renders it. This works, but it misses something fundamental about how React Server Components actually model computation.RSCs do not just move rendering to the server. They make the server a first-class part of the React tree. And the server is exactly where AI inference belongs: close to data, streaming by design, and free from the browser's constraints.In this talk I'll show a concrete architecture where the AI call is the render. A Server Component reaches out to a language model, receives structured output, and streams typed React elements directly to the client with no extra API route, no serialization layer, and no client-side state for the AI response.We'll cover the happy path, the failure modes, and the real-world lessons from building this in production, including hallucination-resistant validation, Suspense boundaries for model latency, and the one architectural mistake that will destroy your Time-to-First-Byte.