惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

aimingoo的专栏
aimingoo的专栏
宝玉的分享
宝玉的分享
博客园 - 【当耐特】
博客园 - 司徒正美
L
LangChain Blog
有赞技术团队
有赞技术团队
大猫的无限游戏
大猫的无限游戏
Stack Overflow Blog
Stack Overflow Blog
Engineering at Meta
Engineering at Meta
U
Unit 42
Microsoft Azure Blog
Microsoft Azure Blog
I
InfoQ
博客园 - 叶小钗
H
Hackread – Cybersecurity News, Data Breaches, AI and More
J
Java Code Geeks
月光博客
月光博客
量子位
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
博客园_首页
Last Week in AI
Last Week in AI
人人都是产品经理
人人都是产品经理
Google DeepMind News
Google DeepMind News
云风的 BLOG
云风的 BLOG
D
DataBreaches.Net

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant
I made an AI that roasts your code (and it's brutally hon...
Antonio Tira · 2026-04-27 · via DEV Community
Cover image for I made an AI that roasts your code (and it's brutally honest)

Antonio Tirado Montes

We've all been there: you push code thinking it's clean, then a senior dev comments "why?" on line 47 and your soul leaves your body.

I built Code Roast to give you that experience before the PR — for free, with AI, in 15 languages.

The brief
Build an AI code reviewer that's funny, honest, and educational.
Bonus: it should hurt a little (so you remember the lesson).

The result
You paste code → the AI returns:

A roast ("This 47-line function violates SRP harder than my New Year's resolutions")
A severity badge (mild → catastrophic) you can share
Actionable fixes (because being mean isn't useful without solutions)
The prompt that makes it work
After 30+ iterations, this is the structure that consistently produces funny + useful:

const prompt = `You are a senior dev reviewing code for a junior. Be:

  • BRUTAL but FAIR (no personal attacks, only the code)
  • SPECIFIC (cite line numbers, name the antipattern)
  • CONSTRUCTIVE (every roast = one concrete fix)
  • FUNNY (one joke max, never at the dev's expense)

Return JSON:
{
"severity": "mild" | "medium" | "spicy" | "catastrophic",
"roast": "<2-3 sentences, the funny part>",
"issues": [{ "line": number, "problem": string, "fix": string }],
"verdict": ""
}

Code to review:
`\
${userCode}
\``;
The key insights:

Constraint produces creativity. "One joke max" forces the model to pick the best one.
Structured output > prose. JSON makes the UI predictable and shareable.
Severity as enum lets us show different badge designs (and a different Kody face per level).
The tech stack (boring on purpose)
Frontend: React + Vite + Tailwind
AI: Gemini 2.5 Flash via Supabase Edge Function
Rate limiting: Postgres rate_limits table, 5 roasts/hour for free, unlimited Pro
Result sharing: dynamic 1200×630 SVG generated in another Edge Function
The whole thing is ~400 lines of code for the feature itself.

What surprised me

  1. Severity badges drive 80% of the shares.
    People love showing off a "Catastrophic 🔥" badge to their friends way more than the actual review. Make the outcome shareable, not the process.

  2. Multilingual roasts hit different.
    A Spanish dev got a roast that included a "madre mía" and tweeted it. Localized humor > translated humor.

  3. Free tier converts.
    About 4% of users hit the free limit and upgrade. The tool itself is the funnel.

Try it (free, no signup needed)
👉 codemasterip.com/code-roast

Paste your worst code. I dare you.

Have you built AI tools that lean into humor? What worked for you? Comments are open.