Despoudis-HI1200×1505 114 KB

Theo Despoudis

Build AI-Enhanced Web Apps is written for developers comfortable with React or Next.js who are ready to integrate large language models into production systems. If you’ve been experimenting with GPT, Claude, Llama, or Gemini and wondering how to move from playground demos to deployed features, this book walks through that process step by step.

Inside, you’ll learn how to:

• Connect LLMs to React and Next.js applications

• Stream responses and handle real-time AI interactions

• Manage conversation state and context in chat apps

• Build RAG systems for knowledge-heavy use cases

• Use LangChain.js for more advanced workflows and reasoning

The projects aren’t toy examples. You’ll build a voice-based interview assistant and a RAG-powered knowledge system that could sit behind an internal company portal. Along the way, the book covers architecture decisions across UI, backend, data processing, API integration, deployment, and scaling.

It also tackles the practical concerns that come up quickly in production: latency, API cost, guardrails for safety and bias, and how to design interactions that make sense for users rather than just showing off a model.

One detail many developers appreciate: the entire stack is JavaScript-based. React for the UI. Next.js for backend integration. The Vercel AI SDK to connect to LLMs. You don’t need a Python or machine learning background to follow along.

If you’re building web apps today, AI features are quickly becoming part of the baseline. This book is meant to help you add them in a way that holds up under real usage.

You can check out the full table of contents and a sample chapter on the Manning site. If you pick it up, I’d love to hear what you’re building with it.

• Full details: Build AI-Enhanced Web Apps - Theo Despoudis

Don’t forget you can get 45% off with your Devtalk discount! Just use the coupon code “devtalk.com” at checkout