AboutAI ToolsCursorClaude CodeLovableBolt.newAI tool reviewsTutorialsVibe codingAI agentsMobile apps with AIAll tutorialsBlogSponsorWork With MeJoin 285K+ Subscribers

Video tutorial · The Metaverse Guy

How I Built a Full-Stack Portfolio Site with GLM 5.2

A step-by-step guide to building a portfolio website with a chatbot and newsletter backend using the KiloCode plugin in VS Code.

GLM 5.2 Is INSANE | Built a Full-stack $10,000 Portfolio Site with AI - Better than Claude Fable 5?
Published
23 June 2026
Length
9 min
Views
3,745 views
On YouTube
GLM 5.2 Is INSANE | Built a Full-stack $10,000 Portfolio Site with AI - Better than Claude Fable 5?

In this article, I walk through building a functional portfolio website entirely with GLM 5.2 and the KiloCode extension in VS Code. I used guided prompts to generate the frontend, integrate a chatbot, and set up a newsletter backend without hand-coding every line. The tutorial shows how to maintain control over the AI-generated output and get a working site.

Setting Up the Environment

I started by installing the KiloCode extension in VS Code. KiloCode connects directly to GLM 5.2, allowing me to prompt the model from within the editor. Once installed, I configured the plugin to use the GLM 5.2 model so I could start generating code immediately.

Generating the Portfolio Frontend

I prompted GLM 5.2 to create a responsive portfolio layout. By providing specific requirements, such as sections for projects, about, and contact, I got a solid foundation in a few iterations. The model handled the CSS and structure, and I only needed to tweak the content.

Integrating a Chatbot

Next, I added a chatbot component. I described the chatbot’s appearance and basic functionality, and GLM 5.2 generated the frontend code. This involved creating a chat window and input field that could later connect to a backend service.

Building the Newsletter Backend

For the newsletter feature, I guided the model to set up a backend endpoint. I specified the data flow—collecting email addresses and storing them—and GLM 5.2 produced the necessary logic. KiloCode helped me scaffold the API routes without writing the boilerplate from scratch.

Controlling and Refining AI Output

Throughout the process, I emphasised the importance of guiding the AI. Instead of accepting all generated code blindly, I reviewed each suggestion, requested changes, and gave targeted prompts. This kept the output aligned with the project’s needs and taught me how to work effectively with AI coding assistants.

Tools used in this video

  • GLM 5.2
  • KiloCode
  • VS Code

Questions

What is GLM 5.2?
GLM 5.2 is an open-weights large language model from Z.AI, designed for strong coding and reasoning performance. I used it through the KiloCode plugin in VS Code.
Do I need to be an expert coder to follow this tutorial?
Basic familiarity with VS Code and web development helps, but the tutorial shows how prompts can generate most of the code. You’ll learn to guide the AI even if you’re not writing everything from scratch.
Can I use this method for other website types?
Yes, the same approach—breaking the project into guided prompts—works for any site. The key is learning how to structure your requests to the model.
What is the KiloCode plugin?
KiloCode is a VS Code extension that lets you interact with various language models, including GLM 5.2, directly from your editor. It simplifies sending prompts and receiving code suggestions.
Video description and links
Build a functional portfolio website using GLM 5.2 and VS Code. See how to integrate a chatbot and newsletter backend step-by-step. This tutorial demonstrates how to leverage the KiloCode plugin to construct a complete portfolio website from scratch. If you are a developer looking to streamline your workflow using AI tools, this guide provides a clear walkthrough of the process. I break down the specific prompts and guided inputs required to make GLM 5.2 generate reliable, functional code for your project. You will learn how to set up a backend architecture and add a newsletter feature without writing every line of code manually. We focus on the practical application of GLM 5.2 within the VS Code environment, ensuring you understand how to maintain control over the output. By the end, you will have a working site and a better grasp of how to use the KiloCode plugin for efficient site development. glm-5.2,glm 5.2,z.ai,open weights model,open source llm,glm vs claude,kimi k2.7,claude opus 4.8,gpt-5.5,deepseek v4,minimax m3,qwen 3.7,moonshot ai,chinese ai models,artificial analysis,intelligence index,design arena,gdpval,best open source ai,local llm,run ai locally,ai coding model,vibe coding,nextjs,three js game,model comparison,open vs closed ai,frontier models Subscribe for weekly AI development breakdowns, and comment below which feature you want to add to your portfolio next.

This article was written from the video's description with the help of AI. Watch the video for the full walkthrough.

Related tutorials