Prakart's Blog

PERT Three-Pont Estimating

Sat, 06 Jun 2026 00:00:00 GMT

Project Management: PERT Three-Point Estimating

If you are estimating the duration of a project or task, PERT stands for Program Evaluation and Review Technique. It calculates a weighted average of three time estimates to account for real-world uncertainty.

The formula is:

$$E = \frac{O + 4M + P}{6}$$

O (Optimistic): The best-case scenario if everything goes perfectly.
M (Most Likely): The expected duration under normal, everyday conditions.
P (Pessimistic): The worst-case scenario if everything goes wrong.

Example

If it takes your team:

3 days in the best case,
5 days normally,
13 days in the worst-case scenario,

then your PERT estimate is:

$$E = \frac{3 + (4 \times 5) + 13}{6} = \frac{36}{6} = 6\text{ days}$$

Zero Supervision

Thu, 30 Apr 2026 00:00:00 GMT

import { Aside } from 'astro-pure/user'

Zero Supervision

Zero supervision refers to AI models and systems that operate, train, or optimize themselves without needing human-annotated data or external guidance.

Recent advancements (2025–2026) focus on inference-time self-evolution and self-generated rewards to achieve state-of-the-art (SOTA) results in coding, reasoning, and text correction.

Key Developments in Zero Supervision (2025–2026)

1. MAS-Zero (Multi-Agent Systems)

A framework that designs multi-agent systems at inference time through self-evolving meta-feedback.

Achieves superior accuracy to manual approaches in coding and reasoning
Does not require labeled training data

2. CEC-Zero (Character Error Correction)

A zero-supervision reinforcement learning framework that allows large language models (LLMs) to correct errors in text by generating their own rewards.

Outperforms supervised models by 10–13 F1 points

3. RLZero (Robot Learning)

Enables zero-shot language-to-behavior policies by "imagining" tasks and grounding them in video-language models.

Removes the need for supervised robot training

4. Absolute Zero (Reasoning)

A paradigm (AZR) where agents propose and solve their own tasks to maximize learning.

Achieves SOTA results in math and code
Requires no external training data

5. AlphaMath "Almost Zero"

Uses Monte Carlo Tree Search (MCTS) to generate self-supervised process rewards.

Matches or exceeds GPT-4-supervised models in math reasoning

Advantages

Cost Efficiency
Eliminates the need for expensive human labor to annotate data
Scalability
Allows models to learn from raw, unlabeled web data or self-simulation
Adaptivity
Enables models to adapt to new problems at inference time

Core Components

Self-Generated Rewards
Models compute their own success metrics (e.g., semantic similarity, consensus)
Self-Evolution
Iteratively improving a model based on its own past performance
Code Execution Verification
Using tools to test and verify answers

Claude Code Login/Register System

Wed, 29 Apr 2026 14:00:00 GMT

Claude Code + Next.js + Better Auth

Overview

สรุปวิธีสร้างระบบ Login/Register โดยใช้:

Next.js (App Router)
Tailwind CSS (Design System)
Better Auth
MySQL
Claude Code (AI-assisted development)

Core Concept

แนวคิดหลักคือ AI-driven development (Vibe Coding)

ใช้ Claude Code + Prompt เพื่อ generate code
ใช้ Skills เพื่อ scaffold ระบบ
ลดการเขียน boilerplate

Architecture

Frontend (Next.js + Tailwind)
        ↓
Auth Client (better-auth/react)
        ↓
API Route (Next.js)
        ↓
Better Auth Core
        ↓
MySQL Database

Step 1: Setup Project

npx create-next-app@latest my-auth-app
cd my-auth-app
npm install

Install Tailwind

npm install tailwindcss
npx tailwindcss init

Step 2: Use Claude Code Prompt

Example Prompt:

Build a login/register system using:
- Next.js App Router
- Tailwind CSS
- Better Auth
- MySQL
- Include register, login, dashboard (protected route)

Claude จะ generate:

Folder structure
API routes
Auth logic
UI pages

Step 3: Setup Better Auth

Install:

npm install better-auth mysql2

Create config:

// lib/auth.ts
import { betterAuth } from "better-auth";
import mysql from "mysql2/promise";

export const auth = betterAuth({
  database: mysql.createPool({
    host: "localhost",
    user: "root",
    password: "",
    database: "auth_db",
  }),
  emailAndPassword: {
    enabled: true,
  },
});

Step 4: API Route

// app/api/auth/[...all]/route.ts
import { auth } from "@/lib/auth";
import { toNextJsHandler } from "better-auth/next-js";

export const { GET, POST } = toNextJsHandler(auth);

Step 5: Auth Client (Frontend)

// lib/auth-client.ts
import { createAuthClient } from "better-auth/react";

export const { signIn, signUp, signOut, useSession } =
  createAuthClient();

Step 6: Register Page

"use client";

import { signUp } from "@/lib/auth-client";

async function handleSubmit(e) {
  e.preventDefault();

  const formData = new FormData(e.currentTarget);

  await signUp.email({
    email: formData.get("email"),
    password: formData.get("password"),
    name: formData.get("name"),
  });
}

Step 7: Login Page

"use client";

import { signIn } from "@/lib/auth-client";

await signIn.email({
  email,
  password,
});

Step 8: Protected Route

const session = await auth.api.getSession({
  headers: await headers(),
});

if (!session) {
  redirect("/login");
}

Step 9: Database (MySQL)

Better Auth จะจัดการ:

user table
session table
account table
verification table

⚡ Key Benefits

1. Claude Code

ลดเวลาเขียน code
Generate ระบบได้เร็ว

2. Better Auth

ไม่ต้องเขียน auth เอง
ลด security risk

3. Tailwind

สร้าง UI ได้เร็ว
Design consistency

Benefits

Traditional Approach

เขียน auth เอง
ใช้เวลานาน
เสี่ยง security bug

Modern Approach

ใช้ Better Auth
ใช้ AI generate code
Focus business logic

Suggested Extensions

Google OAuth Login
Role-based access control (RBAC)
Email verification
Multi-tenant support (SaaS)

Summary

การใช้ Claude Code + Better Auth + Next.js:

ลด development time อย่างมาก
ได้ระบบที่ scalable
เหมาะกับ modern AI-native development workflow

OpenClaw on Hostinger using Claude and Telegram

Wed, 29 Apr 2026 00:00:00 GMT

OpenClaw + Claude + Telegram (Auto AI System)

Overview

สอนติดตั้งระบบ OpenClaw
ใช้ Claude เป็น AI backend
เชื่อมต่อกับ Telegram เพื่อใช้งานจริง
Deploy บน Hostinger (หรือ VPS)

แนวคิดหลัก

ใช้ AI agent ทำงานอัตโนมัติ
ใช้ Telegram เป็น UI interface
ลดการเขียนระบบ backend เอง
ใช้ Claude เป็นสมองของระบบ

Architecture

User → Telegram Bot → OpenClaw → Claude API → Response

ขั้นตอนหลัก

1. เตรียม Server (Hostinger)

ใช้ VPS / hosting ที่รองรับ Node.js
SSH เข้า server

2. ติดตั้ง OpenClaw

clone repo
install dependencies (npm install)
run project

3. ตั้งค่า Claude API

ใส่ API key ของ Claude
config environment (.env)

4. สร้าง Telegram Bot

ใช้ BotFather
ได้ token มาใช้งาน
ใส่ token ใน config

5. เชื่อม Telegram กับ OpenClaw

ตั้ง webhook หรือ polling
bot รับ message → ส่งเข้า AI

6. Run ระบบ

start server
ทดสอบส่งข้อความผ่าน Telegram
AI ตอบกลับแบบ real-time

Use Case

AI Assistant ส่วนตัว
Chatbot สำหรับธุรกิจ
Automation workflow
Dev tools (generate code, debug)

จุดเด่น

setup ง่าย (ไม่กี่คลิก)
ใช้ AI ได้ทันที
ไม่ต้องสร้าง UI เอง
scale ได้ง่าย

ข้อควรระวัง

ต้องจัดการ API key ให้ดี
มีค่าใช้จ่าย Claude API
ต้องดูเรื่อง security (bot + server)

Deployment Summary: Hostinger & GitHub CI/CD

Thu, 23 Apr 2026 00:00:00 GMT

Deployment Summary: Hostinger & GitHub CI/CD

1. Primary Objective

Successfully deploy the local Astro project to a live production environment on Hostinger using automated GitHub-based deployment.

2. Key Technical Activities

Static Build Configuration: Successfully configured the Astro project for static site generation (SSG) to ensure compatibility with standard web hosting.
Production Build Validation: Ran npm run build to generate the production dist/ folder, identifying and resolving dependency and code-level errors.
CI/CD Pipeline Setup: Connected the GitHub repository to Hostinger’s built-in deployment service using SSH Deploy Keys.
Output Directory Mapping: Configured the deployment settings to ensure the server correctly identifies and moves the dist/ folder to the live public_html directory.

3. Problems Encountered & Solutions

Build Crash (Cannot read properties of undefined): * Problem: The production build is "strict" and crashed because the code was attempting to access properties (.toString()) on missing or empty data objects in the projects list.
- Solution: Implemented "Guard Clauses" and "Optional Chaining" (?.) in ProjectSection.astro to safely handle missing project data without crashing the build engine.
Development vs. Production Discrepancy:
- Problem: The site worked on localhost but failed during npm run build.
- Resolution: Recognized that local development environments are "forgiving" (only rendering what is visited), whereas production builds perform a "full coverage" check of all pages.
Deployment Script Error (No output directory found):
- Problem: Hostinger couldn't locate the generated build artifacts.
- Solution: Explicitly mapped the "Output Directory" setting in Hostinger hPanel to match the project's dist folder.

4. Current Status

Deployment: Successfully live.
Workflow: Automated via GitHub push.
Immediate Next Steps: Review the project list on the site; fix the individual project data entries that triggered the original build errors so they can be safely uncommented and displayed.

Here is a summary of the tasks, technical challenges, and solutions from our deployment session today, organized for your notes:

Deployment Summary: Hostinger & GitHub CI/CD

Put simply, 3D rendering is the process of using a computer to generate a 2D image from a digital three-dimensional scene.

To generate an image, specific methodologies and special software and hardware are used. Therefore, we need to understand that 3D rendering is a process—the one that builds the image.

Types of 3D rendering

We can create different types of rendered image; they can be realistic or non-realistic.

A realistic image could be an architectural interior that looks like a photograph, a product-design image such as a piece of furniture, or an automotive rendering of a car. On the other hand, we can create a non-realistic image such as an outline-type diagram or a cartoon-style image with a traditional 2D look. Technically, we can visualize anything we can imagine.

How is 3D rendering used?

3D rendering is an essential technique for many industries including architecture, product design, advertising, video games and visual effects for film, TV and animation.

In design and architecture, renders allow creative people to communicate their ideas in a clear and transparent way. A render gives them the chance to evaluate their proposals, experiment with materials, conduct studies and contextualize their designs in the real world before they are built or manufactured.

For the media and entertainment industries, 3D rendering is fundamental to the creation of sequences and animations that tell stories, whether we’re watching an animated movie, a period drama, or an action sequence with explosions, ships from the future, exotic locales, or extraterrestrial creatures.

Over the past few years, the evolution of computer graphics in these industries has replaced traditional techniques. For example, special effects are being replaced by visual effects, which means stunt people no longer risk their lives in car crashes.

In advertising, I would dare to say that 90% of automotive commercials are CG—or even more. In the architecture industry, many traditional techniques to create representations, such as scale models, have been replaced with photorealistic imagery to ensure we can see exactly how something will look once it’s built.

Accelerating processes, reducing costs and the demand for better quality results have helped technology evolve. Hardware is more powerful than ever and the switch to CG was inevitable.

How is a 3D rendered image generated?

Two pieces of software, with different characteristics, are used to computer-generate images and animations: render engines and game engines. Render engines use a technique called ray tracing, while game engines use a technique called rasterization—and some engines mix both techniques, but we will talk about that later on.

Using MDX

Sun, 01 Jun 2025 00:00:00 GMT

This theme comes with the @astrojs/mdx integration installed and configured in your astro.config.ts config file. If you prefer not to use MDX, you can disable support by removing the integration from your config file.

Why MDX?

MDX is a special flavor of Markdown that supports embedded JavaScript & JSX syntax. This unlocks the ability to mix JavaScript and UI Components into your Markdown content for things like interactive charts or alerts.

If you have existing content authored in MDX, this integration will hopefully make migrating to Astro a breeze.

Example

Here is how you import and use a UI component inside of MDX.
When you open this page in the browser, you should see the clickable button below.

import { Button } from 'astro-pure/user'

Click Me

Markdown 语法支持

Wed, 26 Jul 2023 08:00:00 GMT

基本语法

Markdown 是一种轻量级且易于使用的语法，用于为您的写作设计风格。

标题

文章内容较多时，可以用标题分段：

# 标题 1

## 标题 2

## 大标题

### 小标题

标题预览会打乱文章的结构，所以在此不展示。

粗斜体

_斜体文本_

**粗体文本**

**_粗斜体文本_**

预览：

斜体文本

粗体文本

粗斜体文本

链接

文字链接 [链接名称](http://链接网址)

预览：

文字链接链接名称

行内代码

这是一条 `单行代码`

预览：

这是一条 行内代码

代码块

```js
// calculate fibonacci
function fibonacci(n) {
  if (n <= 1) return 1
  return fibonacci(n - 1) + fibonacci(n - 2)
}
```

预览：

// calculate fibonacci
function fibonacci(n) {
  if (n <= 1) return 1
  return fibonacci(n - 1) + fibonacci(n - 2)
}

当前使用 shiki 作为代码高亮插件，支持的语言请参考 shiki / languages。

行内公式

这是一条行内公式 $e^{i\pi} + 1 = 0$

预览：

这是一条行内公式 $e^{i\pi} + 1 = 0$

公式块

$$
\hat{f}(\xi) = \int_{-\infty}^{\infty} f(x) e^{-2\pi i x \xi} \, dx
$$

预览：

$$ \hat{f}(\xi) = \int_{-\infty}^{\infty} f(x) e^{-2\pi i x \xi} , dx $$

当前使用 KaTeX 作为数学公式插件，支持的语法请参考 KaTeX Supported Functions。

图片

![CWorld](https://gravatar.loli.net/avatar/1ffe42aa45a6b1444a786b1f32dfa8aa?s=200)

预览：

删除线

~~删除线~~

预览：

~~删除线~~

列表

普通无序列表

- 1
- 2
- 3

预览：

普通有序列表

1. GPT-4
2. Claude Opus
3. LLaMa

预览：

GPT-4
Claude Opus
LLaMa

列表里可以继续嵌套语法

引用

> 枪响，雷鸣，剑起。繁花血景。

预览：

枪响，雷鸣，剑起。繁花血景。

引用里也可以继续嵌套语法。

换行

markdown 分段落是需要空一行的。

如果不空行
就会在一段

第一段

第二段

预览：

如果不空行就会在一段

第一段

第二段

分隔符

如果你有写分割线的习惯，可以新起一行输入三个减号--- 或者星号 ***。当前后都有段落时，请空出一行：

---

预览：

高级技巧

行内 HTML 元素

目前只支持部分段内 HTML 元素效果，包括 <kdb>  ，如

键位显示

使用 <kbd>Ctrl</kbd> + <kbd>Alt</kbd> + <kbd>Del</kbd> 重启电脑

预览：

使用 Ctrl + Alt + Del 重启电脑

粗斜体

<b> Markdown 在此处同样适用，如 _加粗_ </b>

预览：

Markdown 在此处同样适用，如加粗

其他 HTML 写法

折叠块

<details><summary>点击展开</summary>它被隐藏了</details>

预览：

表格

| 表头1 | 表头2 |
| ----- | ----- |
| 内容1 | 内容2 |

预览：

| 表头1 | 表头2 | | ----- | ----- | | 内容1 | 内容2 |

注释

在引用的地方使用 [^注释] 来添加注释。

然后在文档的结尾，添加注释的内容（会默认于文章结尾渲染之）。

[^注释]: 这里是注释的内容

预览：

在引用的地方使用 ^注释来添加注释。

然后在文档的结尾，添加注释的内容（会默认于文章结尾渲染之）。

To-Do 列表

- [ ] 未完成的任务
- [x] 已完成的任务

预览：

[ ] 未完成的任务
[x] 已完成的任务

符号转义

如果你的描述中需要用到 markdown 的符号，比如 _ # * 等，但又不想它被转义，这时候可以在这些符号前加反斜杠，如 \_ \# \* 进行避免。

\_不想这里的文本变斜体\_

\*\*不想这里的文本被加粗\*\*

预览：

_不想这里的文本变斜体_

**不想这里的文本被加粗**

内嵌 Astro 组件

See User Components and Advanced Components for details.

Prakart's Blog

PERT Three-Pont Estimating

Project Management: PERT Three-Point Estimating

The formula is:

Example

Zero Supervision

Zero Supervision

Key Developments in Zero Supervision (2025–2026)

1. MAS-Zero (Multi-Agent Systems)

2. CEC-Zero (Character Error Correction)

3. RLZero (Robot Learning)

4. Absolute Zero (Reasoning)

5. AlphaMath "Almost Zero"

Advantages

Core Components

Claude Code Login/Register System

Claude Code + Next.js + Better Auth

Overview

Core Concept

Architecture

Step 1: Setup Project

Install Tailwind

Step 2: Use Claude Code Prompt

Step 3: Setup Better Auth

Step 4: API Route

Step 5: Auth Client (Frontend)

Step 6: Register Page

Step 7: Login Page

Step 8: Protected Route

Step 9: Database (MySQL)

⚡ Key Benefits

1. Claude Code

2. Better Auth

3. Tailwind

Benefits

Traditional Approach

Modern Approach

Suggested Extensions

Summary

OpenClaw on Hostinger using Claude and Telegram

OpenClaw + Claude + Telegram (Auto AI System)

Overview

แนวคิดหลัก

Architecture

ขั้นตอนหลัก

1. เตรียม Server (Hostinger)

2. ติดตั้ง OpenClaw

3. ตั้งค่า Claude API

4. สร้าง Telegram Bot

5. เชื่อม Telegram กับ OpenClaw

6. Run ระบบ

Use Case

จุดเด่น

ข้อควรระวัง

Deployment Summary: Hostinger & GitHub CI/CD

Deployment Summary: Hostinger & GitHub CI/CD

Deployment Summary: Hostinger & GitHub CI/CD

Types of 3D rendering

How is 3D rendering used?

How is a 3D rendered image generated?

Using MDX

Why MDX?

Example

More Links

Markdown 语法支持

基本语法

标题

粗斜体

链接

行内代码

代码块

行内公式

公式块

图片

删除线

列表

引用

换行

分隔符

高级技巧