Danielle Hoopes

1. 01

   Fig. 01

   Vector Databases: Why Your AI App Actually Needs One (2026 Explainer)

   6 pcs · 7 min

   Your database can find rows containing "refund." It can't find the ticket that says "I want my money back." Here's why search-by-meaning breaks regular databases, what vector databases actually do, and when pgvector is all you need.

   • 01 Embeddings in 60 Seconds
   • 02 Why Your Regular Database Can't Do This
   • 03 What a Vector Database Actually Does
   • 04 Why Not Just Use a Bigger Context Window?
   • 05 When You Don't Need a Dedicated One
   • 06 The Options in 2026
2. 02

   Fig. 02

   Tokenmaxxing: AI Tokens Are Becoming the New Productivity Metric (And Why That's Half-Right)

   6 pcs · 6 min

   A developer in Iceland built a desk gadget that shows AI token usage as pixel art. It went viral. Here's why "tokenmaxxing" is a status symbol, why tokens are a decent proxy for AI fluency, and why the metric breaks the second your manager tracks it.

   • 01 What the Clawdmeter Actually Is
   • 02 Why Tokens Are Suddenly a Status Symbol
   • 03 The Good Case for Tokens as a Metric
   • 04 The Bad Case (And It's a Real One)
   • 05 What Good AI Usage Actually Looks Like
   • 06 What Leaders Should Be Asking Right Now
3. 03

   Fig. 03

   Where Are We With Claude Mythos? (And Do You Actually Need It?)

   6 pcs · 11 min

   Anthropic's 10-trillion-parameter Mythos model is real, locked behind Project Glasswing, and you cannot use it. Here's the actual status, an honest answer on whether you need it, and what to ship with right now while you wait.

   • 01 What We Actually Know About Mythos
   • 02 Why Anthropic Locked It Away
   • 03 Do You Actually Need It?
   • 04 What to Use Right Now Instead
   • 05 When You Might Genuinely Need to Wait
   • 06 What the Lockup Signals for the Next 12 Months
4. 04

   Fig. 04

   The AI Price War Is Getting Ridiculous (And Developers Win)

   7 pcs · 8 min

   AI API prices dropped 98% in two years. GPT-5.4 at $2.50/M tokens, Gemini Flash-Lite at $0.25/M. Here's the full breakdown, which model to use for what, and why indie devs should be celebrating.

   • 01 What Happened This Month
   • 02 The Price Collapse Timeline
   • 03 Head-to-Head: Who Costs What (March 2026)
   • 04 Which Model Should You Actually Use?
   • 05 Why Prices Are Crashing
   • 06 What This Means for Indie Developers
   • 07 The Catch (There's Always a Catch)
5. 05

   Fig. 05

   How I Made Claude Code Enforce OWASP Rules (So I Don't Have To)

   7 pcs · 10 min

   Claude Code skills can automatically enforce security patterns, catch OWASP violations, and flag accessibility issues before they ship. Here's the security skill I built for my team.

   • 01 The Problem With Security Knowledge
   • 02 What Are Skills (30-Second Version)
   • 03 Your Security Rules Here
   • 04 The Security Skill I Built
   • 05 OWASP Top 10 in Practice
   • 06 Accessibility Is Security Too
   • 07 Sharing This With Your Team
6. 06

   Fig. 06

   Claude Skills: What They Are, Why You Need Them, and How to Set Them Up

   10 pcs · 12 min

   Skills are reusable instruction packs that make Claude actually follow your standards. Here's how to find, install, and build your own in 10 minutes.

   • 01 The Problem Skills Solve
   • 02 What Are Claude Skills?
   • 03 Instructions
   • 04 Installing Skills in 2 Minutes
   • 05 Where Skills Live
   • 06 Writing Your Own Skill
   • 07 PR Review Checklist
   • 08 Real Example: Accessibility
   • 09 The skills.sh Marketplace
   • 10 Should You Bother?
7. 07

   Fig. 07

   My Actual Dev Setup: Google Antigravity + Claude Code

   6 pcs · 9 min

   Everyone keeps asking: Antigravity or Claude Code? My answer: both. Here's how I use Google's agent-first IDE alongside Anthropic's terminal-first CLI and why I stopped trying to pick just one.

   • 01 What Is Google Antigravity?
   • 02 What Is Claude Code?
   • 03 How I Actually Use Them Together
   • 04 What Each One Is Better At
   • 05 The Honest Downsides
   • 06 Should You Try This?
8. 08

   Fig. 08

   Cursor vs. Windsurf vs. Claude Code: The AI Coding Editor War (Honest Guide)

   7 pcs · 10 min

   The $29B hype machine, the 72-hour acquisition soap opera, and the terminal purist. An honest comparison of the three AI coding tools everyone is arguing about.

   • 01 The Three Contenders
   • 02 Cursor: The $29 Billion Hype Machine
   • 03 Windsurf: The 72-Hour Acquisition Saga
   • 04 Claude Code: The Terminal Purist
   • 05 The Real Comparison
   • 06 Which One Should You Use?
   • 07 The Uncomfortable Truth
9. 09

   Fig. 09

   OpenClaw's Dirty Secret: It Can Burn Hundreds of Dollars a Day

   6 pcs · 9 min

   Nobody in the hype cycle is talking about what OpenClaw actually costs. Real horror stories, why the architecture burns tokens, and how to not go broke running it.

   • 01 The Horror Stories Are Real
   • 02 Why It Burns So Many Tokens
   • 03 The Real Math
   • 04 How to Not Go Broke
   • 05 What About the Hosted Versions?
   • 06 Should You Even Bother?
10. 10

    Fig. 10

    MCP Tutorial: Build AI Agents That Actually Do Things (2026 Guide)

    11 pcs · 11 min

    Learn Model Context Protocol (MCP) - the "USB-C for AI" that OpenAI, Microsoft, and Anthropic all adopted. Step-by-step tutorial to build your first AI agent.

    • 01 What is MCP?
    • 02 Why Should You Care?
    • 03 Prerequisites
    • 04 Step 1: Install the MCP SDK
    • 05 Step 2: Create a Simple MCP Server
    • 06 Step 3: Connect Claude to Your Server
    • 07 Security Considerations
    • 08 Real-World MCP Servers
    • 09 Hosting Your MCP Agent
    • 10 What's Next?
    • 11 Resources
11. 11

    Fig. 11

    What is Clawdbot? Do You Need It? (The No-Hype Explainer)

    7 pcs · 7 min

    Your group chat won't shut up about Clawdbot. Here's what it actually is, whether you need it, and why 1,800+ people left theirs wide open to the internet.

    • 01 What Is It, Actually?
    • 02 Why Is Everyone Freaking Out?
    • 03 Do You Need It?
    • 04 The Security Thing Nobody Wants to Talk About
    • 05 The 2-Minute Setup Check
    • 06 What Should You Actually Do?
    • 07 The Bottom Line
12. 12

    Fig. 12

    Vibe Check: The Linter That Catches AI Slop Before It Ships

    8 pcs · 8 min

    Introducing vibe-check, a CLI tool that catches common AI-generated code issues like any types, lazy comments, generic variable names, console spam, and empty catch blocks.

    • 01 What is AI Slop?
    • 02 What vibe-check Catches
    • 03 Installation & Usage
    • 04 Configuration
    • 05 CI/CD Integration
    • 06 Why I Built This
    • 07 Try It Out
    • 08 Support the Project
13. 13

    Fig. 13

    Clawdbot (OpenClaw): Complete Setup Guide + Security Concerns You Need to Know

    10 pcs · 10 min

    Everything about Clawdbot/OpenClaw - the viral AI assistant with 100k+ GitHub stars. Includes step-by-step setup, security risks (1,800+ exposed instances), and how to protect yourself.

    • 01 What Is Clawdbot/OpenClaw?
    • 02 Why Did It Go Viral?
    • 03 Key Features
    • 04 The Name Change Drama
    • 05 How to Get Started
    • 06 Running Fully Local (No Cloud APIs)
    • 07 What Can It Actually Do?
    • 08 The Bigger Picture
    • 09 Should You Use It?
    • 10 Resources

• 01 TypeScript · Node.js · CLI

  vibe-check ↗

  The Grammarly for AI code. Catches AI slop before it ships — any types, lazy comments, generic names, console spam, and empty catch blocks.

  Source

• 02 React · Claude AI · Firebase · PWA

  grocery-run ↗

  Snap a recipe screenshot, get a grocery list. Claude Vision extracts ingredients from TikTok and Instagram recipes, groups them by aisle, and syncs across devices. Installable PWA.

• 03 React · Claude AI · Claude Code · Firebase

  house-hunter ↗

  AI-powered property tracker for Colorado Springs. Includes custom Claude Code skills for neighborhood research workflows.

  Source

• 04 React · Claude AI · Tailwind

  fridge-check ↗

  AI recipe suggestions from your fridge contents. Shows the environmental impact — water, energy, CO₂ — of each AI request. Eco mode uses 90% less resources.

  Source

• 05 React · Claude AI · Tailwind

  quickmock ↗

  Visual API mock builder with AI-powered data generation. Create mock endpoints, configure responses, and generate realistic test data with Claude.

  Source

• 06 Angular · TypeScript · Claude AI · RxJS

  storage-finder ↗

  Self-storage marketplace with a Claude AI chatbot. Angular 17 and RxJS, built to WCAG 2.0 accessibility standards.

  Source

Colophon

Background

Full stack developer building enterprise applications. Started in .NET/C#, now equally comfortable in React, Python, and modern JavaScript ecosystems. Experienced with large-scale systems, security-conscious architecture, and cross-functional team collaboration.

Current focus

AI integration, developer tooling, and enterprise automation. I build AI-powered workflows that make teams faster without sacrificing reliability or security.

Philosophy

I solve problems first, pick frameworks second. Clean code matters, but shipping matters more. Accessibility isn’t optional. Documentation is a love letter to future you.

Off duty

When I’m not coding, I’m probably at a concert or exploring Colorado.