Skip to main content

Video podcast

Beyond ChatGPT: What Generative AI Really Is

Duration: 55:34 · With Mangat Rai

From Collin Software Engineering (opens in a new tab)

Loading the video connects to YouTube. Player privacy details

Everybody is using ChatGPT, but what is really happening behind it? In this talk at the Collin College Frisco campus on November 13, 2026, Mangat explains in simple words what generative AI is and the main building blocks behind it. No AI background needed.

What the talk covers​

The session was for students who use ChatGPT every day and want a peek behind it. It covers:

  • What generative AI is, in plain language.
  • The main building blocks: large language models (LLMs), retrieval-augmented generation (RAG), vector databases, AI agents, and AI coding tools.
  • How companies are using AI in real projects today.
  • How AI is changing entry-level jobs in tech.
  • What students can start doing now to be ready.

The session closed with open Q&A, but the students' questions came one on one after the talk, so the recording does not include them.

About the speaker​

Mangat Rai is Director, Solutions Engineering, with over 20 years of experience in the technology industry. He started his career building Java and integration solutions for large enterprises, and later moved into solutions engineering, where he worked with companies from the first discovery call all the way to production go-live. For the last 4 years his focus has been on AI, including generative AI, RAG, and agentic AI. Outside of work he mentors startups and students who want to grow their career in tech.

Where to go next​

If the talk made you curious, these free Few-Shot Academy pages pick up where it leaves off:

  • Foundations: how LLMs work, from basic concepts to a small RAG app that runs on your own laptop.
  • Intermediate: retrieval, prompting, tools, agents, memory, and evaluation.
  • Vector Databases: what a vector database does and why RAG depends on one.
  • Career tracks: five roles that build, operate, sell, or manage AI systems, and what each one asks for.

Comments

Comments are provided by Giscus and GitHub. Loading them connects your browser to GitHub, and GitHub's privacy terms apply.

Privacy details