

#267 - Step-by-Step: Build a Real AI Project with Next.js & RAG
Aug 25, 2025
Explore the intriguing world of AI engineering and discover what it truly means to build real projects. Learn how to develop a Retrieval Augmented Generation app using Next.js and TypeScript. The discussion covers techniques like web scraping, storing content in vector databases, and reducing AI hallucinations for reliable outputs. The host shares insights from their experience in AI startups, highlighting the skills that employers are currently seeking in the ever-evolving tech landscape.
AI Snips
Chapters
Transcript
Episode notes
Title Means Less Than Projects
- The title 'AI engineer' is overloaded and vague but building practical projects matters more.
- Real value comes from projects that show you can apply AI to solve real problems.
From Layoff To Building RAG In Startup
- Brian recounts getting laid off and joining a tiny AI startup where he built production AI apps.
- He worked closely with a former Amazon principal engineer to deliver a Retrieval Augmented Generation system.
RAG Reduces Hallucination
- Retrieval Augmented Generation (RAG) attaches external documents to LLM prompts to reduce hallucination.
- Feeding LLMs relevant context makes them accurate and useful for domain-specific tasks.