Hammad Haqqani - Blog

The official docs tell you how Claude Code works when it works.

Every AI tone changer works the same way: you paste your text, it goes to a language model somewhere, and a rewritten version comes back.

Every Terraform incident postmortem contains the same sentence: "the plan was reviewed and approved."

The highest-leverage file in agent tooling right now is a markdown file with three lines of YAML on top.

Paste this into any Claude Code session: /context.

The SQL assistant on my site had a bug.

The most counterintuitive finding in agent tooling this year: teams measured LLM-generated AGENTS.md files making their agents worse, with lower task success and higher cost.

Two engineers vibe-code on a Friday afternoon.

Ask Claude Code to find where rate limiting is implemented in a 400-file codebase and watch your context window.

Open a fresh Claude Code session in one of my client repos and run /context.

Run /context in your next Claude Code session before you type anything.

A client called me at 11:47 PM on a Tuesday.

In March I shipped Awesome DevOps AI, a curated list of AI tools, agents, and resources built for DevOps, SRE, and Platform Engineering.

Every API price comparison you read uses a price-per-million-tokens number.

A Postgres database that performs well in development and falls over in production is almost always missing an index.

NAT Gateway is the single largest hidden line item on most AWS bills under 50K a month.

Half my consulting calls this quarter started the same way: "We have been on the OpenAI SDK for two years.

MCP (Model Context Protocol) is the standard that lets Claude Code and Claude Desktop talk to external tools, databases, browsers, and APIs.

CLAUDE.md is the agent's onboarding doc.

GitHub Actions billing has four moving parts most teams underestimate: minutes, OS multipliers, matrix expansion, and artifact storage.

Cron expressions are the most ubiquitous and least readable string in DevOps.

A Claude Code agent without hooks is an intern with sudo.

A founder I know left a Claude Code agent running on a refactoring task at 11 PM.

I ran the same 30 production-shaped tasks through GPT-5 (via Codex CLI) and Claude Opus 4.7 (via Claude Code, 1M context, native plan mode).

In the last 90 days I sat down with 12 engineering teams who had rolled out Claude Code, Cursor, or both at company scale.

Over the last 30 days I ran Claude Code as the primary pair-programmer across seven concurrent projects spanning consumer mobile, backend services, trading automation, personal web, networking,...

In the last few months, system prompts from every major AI tool have leaked onto GitHub.

Vibe coding changed how I build software.

There are hundreds of Claude Code plugins now.

What if Claude Code wasn't just a coding assistant, but an entire agile team?

Two months ago I published a post about the 71 free browser-based AI and DevOps tools I built.

Last month, a Fortune 500 company's internal AI chatbot started leaking employee salary data.

You're halfway through pasting your 50-page architecture document into ChatGPT when it stops responding.

"Should I fine-tune or use RAG?"

Over the past 6 months, I've watched the DevOps AI landscape explode from a handful of experimental tools to an overwhelming ecosystem of 500+ products, agents, and frameworks.

You've probably tried using ChatGPT or Claude to write Terraform.

OpenClaw is an open-source personal AI assistant with over 224,000 GitHub stars.

Every week I watch developers paste production secrets into random websites to decode a JWT, estimate token costs by doing math in a spreadsheet, and pay $20/month for a prompt testing tool that...

I've spent over a decade building and scaling systems, leading engineering teams, and staying ahead of every technological shift this industry has thrown at us.

I have been using Claude Code every day for over a year to build infrastructure, write Terraform, debug Kubernetes manifests, and ship production code faster than I ever thought possible.

DevOps in 2026 is unrecognizable from five years ago.

The intersection of artificial intelligence and cloud infrastructure is no longer a futuristic concept -- it's the new reality of DevOps engineering.

Operations controls are the backbone of any secure, well-governed AWS environment.

What if your infrastructure could detect problems, diagnose root causes, and fix itself -- all before an engineer even gets paged?

You've started using Claude Code for infrastructure work.

Technical interviews follow patterns.

The AWS CLI is the foundation of any cloud engineer's toolkit.

Retrieval Augmented Generation (RAG) is one of the most practical AI patterns for real-world applications.

AIOps -- Artificial Intelligence for IT Operations -- applies machine learning to the massive volumes of data generated by modern infrastructure.

After 10+ years of building on AWS and Azure across financial services, government, and enterprise, here are the real lessons I've learned about cloud transformation -- what works, what doesn't, and...