Welcome to claudefan.dev: Why I've created this website

I’ve been a software engineer for years, building everything from distributed systems to enterprise applications. But in the past year, something changed. The way I write code, solve problems, and think about software development has fundamentally shifted — and it’s all because of AI.

This website is my attempt to document that journey and share what I learn along the way.

Why I Created This Website

Let me be honest: I’m a Claude fan.

Not in a fanboy way, but in a “this tool genuinely makes me a better developer” way. I’ve tried GPT, Gemini, and various other models. They’re all impressive. But Claude has become my go-to for coding, problem-solving, and even thinking through complex architectural decisions.

When I started using Claude seriously — especially Claude Code and the newer Opus models — I realized I was learning faster than ever. Not just about AI, but about software development itself. Claude doesn’t just give me answers; it teaches me patterns, challenges my assumptions, and helps me see problems from angles I hadn’t considered.

I wanted a place to document these learnings. Not for clicks or clout, but because writing forces clarity. And if my notes help someone else along the way, even better.

What You’ll Find Here

This blog will focus on practical, hands-on content for developers working with AI:

• Tutorials — Step-by-step guides for integrating AI into real workflows (like self-hosting n8n, building automation pipelines, or setting up development environments)
• Claude Tips — Prompting techniques, use cases, and workflows that actually work
• AI Tools & Infrastructure — Reviews and setup guides for tools in the AI ecosystem
• Learning Notes — Raw notes from my experiments, including what failed and why

I’m not interested in writing theoretical think-pieces about AGI timelines. This blog is for builders — people who want to ship things and learn by doing.

Why Claude? Why Now?

We’re at an inflection point in AI development. The models released in 2025 aren’t just incrementally better — they’re qualitatively different.

Claude Opus 4.5, released just weeks ago, can sustain complex coding tasks for hours. It scored higher than any human candidate on Anthropic’s internal engineering exam. It uses 76% fewer tokens while matching previous state-of-the-art performance.

But benchmarks don’t capture what actually matters. What matters is this: I can now describe a complex refactoring task in plain English and watch Claude execute it across multiple files, maintaining context, fixing tests, and explaining its reasoning along the way.

That’s not autocomplete. That’s a genuine collaborator.

Anthropic’s approach to safety (Constitutional AI) also resonates with me. As someone building tools that others will use, I care about deploying AI responsibly. Claude’s design philosophy — helpful, harmless, honest — aligns with how I think about building software.

The AI Learning Curve is Real

Here’s something I’ve noticed: most developers are either ignoring AI entirely or using it superficially.

The ones ignoring it are falling behind. The ones using it superficially — treating it like a fancy Stack Overflow — are missing the real opportunity.

The developers who are pulling ahead are the ones who’ve invested time in learning how to work with AI. They understand prompting, context management, and when to trust (or not trust) model outputs. They’ve developed intuition for what AI is good at and where it falls short.

This skill gap is widening every month. And it’s not something you can cram in a weekend. It requires consistent practice and experimentation.

That’s another reason I’m writing this blog. Documenting my experiments forces me to be intentional about learning. And sharing publicly keeps me accountable.

Who This Blog Is For

If you’re a developer who:

• Wants to integrate AI into your daily workflow
• Is curious about Claude but hasn’t gone deep yet
• Prefers practical tutorials over theoretical discussions
• Learns by building and breaking things

Then this blog is for you.

I’m not an AI researcher. I’m not building foundation models. I’m a working developer figuring out how to use these tools effectively. Everything I write comes from that perspective.

What’s Coming Next

I’ve already started working on a few posts:

• How I Used Claude to Build claudefan.dev — A deep dive into building a modern tech blog using Claude AI as my pair programming partner
• How to Self-Host n8n for FREE using Render and Supabase — A complete guide to running your own automation platform
• Building AI-Powered Workflows — Practical automation examples using Claude + n8n

I’ll also share shorter notes on prompting techniques, tool discoveries, and lessons from failed experiments.

Let’s Build Together

AI is moving fast. Too fast to learn alone.

If you’re on a similar journey — figuring out how to build effectively with AI — I’d love to hear from you. Drop a comment, reach out on Twitter, or just follow along.

The best time to start learning AI was two years ago. The second best time is now.

Let’s build.

---

— Mohan, builder and Claude enthusiast