AI Didn’t Kill the Web, It Moved in! — Olivier Leplus (AWS) & Yohan Lasorsa (Microsoft)

2026-04-10 · Source: AI Engineer · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems · Depth: Intermediate, extended

Summary

This session, led by Yohan and Olivier, explores the profound impact of recent AI innovations on web development, emphasizing AI's integration across the entire web application lifecycle, from coding and debugging to performance tuning and native browser integration. The discussion highlights the emergence of AI coding agents that utilize "skills" (lightweight, text-based plugins) to add domain expertise and enable repeatable workflows, demonstrated by implementing a contact page and automating testing with Playwright CLI. It also covers AI-powered debugging and performance analysis via Chrome DevTools MCP, allowing agents to control browser functionalities like network throttling and Lighthouse audits. Furthermore, the session introduces experimental local AI APIs within browsers for tasks such as summarization, proofreading, and multimodal prompting, which operate client-side without external API calls. Finally, it addresses adapting web applications for AI agents through proposals like `LLM.txt` for content discovery and Web MCP for exposing web page functionalities as callable tools.

Key takeaway

For web developers building modern applications, you should begin experimenting with AI integration now. Utilize AI coding agents with custom skills for repeatable workflows and explore AI-assisted debugging in browser dev tools. Critically, prepare your web applications for AI agent consumption by implementing `LLM.txt` and considering the highly experimental Web MCP to expose functionalities as callable tools, ensuring your sites remain discoverable and usable in the evolving agentic web.

Key insights

AI is transforming web development across the entire lifecycle, from coding to agent-centric web app design.

Principles

• AI agents benefit from domain-specific "skills."
• Local AI models enable client-side processing.
• Web apps must adapt for AI agent consumption.

Method

Develop web applications using AI coding agents with custom skills, debug and optimize with AI-controlled browser dev tools, and integrate local AI APIs for client-side functionalities like summarization and multimodal prompting.

In practice

• Implement `LLM.txt` for AI agent content discovery.
• Expose web forms as Web MCP tools for agents.
• Enable AI assistance in browser dev tools for debugging.

Topics

• AI in Web Development
• AI Coding Agents
• Browser AI APIs
• Web MCP
• LLM.txt

Best for: AI Engineer, Software Engineer

Related on AIssential

• The agent-ready web: Simplify user actions with WebMCP — Tara Agyemang, Google — Web MCP transforms websites into structured APIs for AI agents, simplifying complex user actions.
• Helping people write code again — LLM-assisted coding empowers experienced individuals to resume programming by reducing time barriers.
• Agents should interview you — AI agents can significantly accelerate development by interviewing users for requirements and automating code generation and testing.
• IDE for AI Browser Agents — Advanced tools and frameworks are emerging to enhance AI agent development, browser automation, and memory management.
• From idea to AI app: Creating intelligent research assistants with Strands — Strands Agents and Kiro simplify AI agent development, enabling complex applications with minimal code and LLM-driven reasoning.
• Perhaps not Boring Technology after all — Modern coding agents can effectively utilize novel or private tools by consuming documentation and iterating.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI Engineer.