// 6 articles on AI by Faez Tgh — practical React, Next.js, TypeScript and web-performance write-ups.
Search stopped being ten blue links. Google's AI Overviews, ChatGPT, and Perplexity now synthesize answers and cite a handful of sources. GEO is the practice of becoming one of those sources — and it rewards the same fundamentals good SEO always did, plus a few new ones.
Bolting an LLM onto a Next.js app is easy to demo and easy to get wrong. Here's the shape of a streaming chat feature that survives contact with real users — plus the rate-limiting, error, and cost details the tutorials skip.
Asking a model to 'reply in JSON' and hoping is how you end up parsing prose at 2am. Modern models expose schema-constrained output and tool calling that guarantee the shape — the trick is choosing the right one and validating anyway.
An LLM only knows what it was trained on and what you paste into the prompt. RAG is how you give it your own knowledge — your docs, your data — without retraining anything. Here's the pipeline, end to end.
Every AI tool used to invent its own way of plugging into your data. MCP is the attempt to standardize that connection — one protocol, many clients, reusable servers. Here's the mental model, minus the hype.
YouTube's built-in speed controls barely scratch the surface. This open-source Chrome extension brings precise slider control, per-channel automation rules, multi-profile presets, and even reverse playback to your browser toolbar.