Let AI do everything?
You'll understand nothing.
People who understand AI will be in demand — because it's hard. I teach the concepts behind it.
6+ Years Industry Experience · AWS & Azure Certified · Fullstack AI Engineer
[ Click through ]
Looks great.
AI built your chatbot. The demo works.
AI is a powerful tool.
Used right, it multiplies you. Used wrong, it replaces the part of you that companies actually pay for.
Two directions
Outsource everything
- Think less yourself
- Question less
- Understand less
→ Obsolete. You stop delivering value.
Understand and question
- Grasp the concepts
- Think critically
- Use tools on purpose
→ More in demand than ever.
Understand instead of outsource — start with free Basic membership.
How?
Five assets that build understanding.
You don't win by using less AI. You win by understanding more. Each asset pulls you from prompting to engineering — and they reinforce each other.
From dependency to depth.
Courses & Blog
Concepts, not hacks
Learn why systems work — so you can explain what you built in any interview.
Starter Kits
Real boilerplates
Build production systems yourself — not demos you prompted into existence.
Community
Direct access
Gold includes a monthly live call — bring one concrete question about what you are building.
AI-in-a-Shell
Your AI tutor
AI that teaches and challenges you — instead of doing the thinking for you.
Knowledge Graph
Your second brain
Connect concepts across everything you learn — so understanding compounds.
6 years in industry. Then I made the transformation myself.
I was afraid of becoming obsolete — so I went from Cloud Architect to AI Engineer. I show you exactly what matters and what's actually in demand.
Today I have a well-paid AI engineer job — but only because I built and understood new things on the side of my day job. I want to help you do the same.
Engineers get paid to solve problems. How are you going to do that if you don't understand the concepts?
Concepts learned — not outsourced
AWS
AI Practitioner
Early Adopter · 2025
Microsoft
Azure AI Fundamentals
2024
Hugging Face
AI Agents Fundamentals
2025
AWS
Solutions Architect
2022
AWS
Developer Associate
2022
See how I teach — for free.
Others sell Masterclass-level content and MCP courses. Mine are free — Lesson 1 needs no account, the rest unlocks with free Basic.
Try first, then commit. Basic membership is free — one login, your email in the system, full depth unlocked.
Claude Code Masterclass
Claude Code is a power engine — the full masterclass is free. Join Basic membership (free) — one login, your email is in the system.
Join Basic & start
MCP Foundations
MCP connects AI to your applications. Lesson 1 is open now — no account. Full course is free with Basic membership — same one login.
Try Lesson 1 freeMy mental model — in six parts.
Six articles that build my AI engineering mental model — principles, not prompt tricks. Free to read, no account needed.
The AI Engineering Mental Model
AI engineering isn't ML engineering and it isn't "call the API with a good prompt." It's building reliable applications on top of probabilistic foundation models — and the most important shift is that LLMs are not functions.
Models as Probabilistic Infrastructure
The model you call today was trained on different comparison data than the one you called three months ago. Post-training, sampling, and structured outputs — and why the same prompt returns different answers.
Evaluation First
Evaluation is the hardest problem in AI engineering and where most teams invest the least. Three methods — exact match, AI-as-judge, comparative — and how to build an eval pipeline that makes quality measurable.
Prompt and Context Engineering
A prompt is a specification, not a command. System vs user prompts, in-context learning, chain-of-thought, the context window as a constrained resource, and defensive prompt engineering against injection and extraction.
Production AI Architecture
A demo is a solved problem. Shipping something that works reliably for a thousand users a day is a different engineering problem. Five production patterns — guardrails, context enhancement, model routing, caching, agents — and the monitoring loop that makes AI systems improve over time.
RAG vs Finetuning — The Decision Framework
The most common question in AI development, answered with a decision tree. When to use RAG, when to finetune, why most finetunes don't work — and the three conditions that justify trying. PEFT and LoRA as the practical entry point.
Latest blog posts
- Jul 14, 20263 minai-agents
Vercel Labs' Agent Browser: The Missing Feedback Loop for Coding Agents
Coding agents can write code, run it, even write tests — but they've never been able to check their own work in a browser. Agent Browser, a new Rust CLI from Vercel Labs, closes that loop.
- Jul 14, 20263 minobservability
Monitoring AI Agents Locally: Langfuse + OpenTelemetry, No Cloud Account Required
Your agent works in a demo. Then it ships, something breaks in production, and you have zero visibility into what it actually did. Here's how to trace every run — self-hosted, on your own machine.
- Jul 14, 20263 minclaude-code
Claude Code Agent Teams: When Sub-Agents Aren't Enough
A sub-agent reports back once and disappears. An Agent Team stays in the room — teammates share a task list, message each other, and challenge each other's findings before you see the result.
AI engineering, weekly.
Weekly patterns by email — or join Basic free for courses and progress.
Blog is open. Basic membership is free — one login unlocks the rest.