Agentic coding · by Ray Amjad
Fast is easy. Right is the job.
Writing the code was never the hard part. Saying exactly what you want, and proving you got it, is. These classes are that process, built around a coding agent.
- Students taught
- Video lessons
- YouTube subscribers
Classes taken by students working at…
A fast agent on a codebase it cannot read is just faster mess.
Coding agents write more code per hour than any human ever has. Left unguided they also make the codebase harder to read, and the harder it is to read, the worse the next run performs. That loop runs in one direction.
Nothing about that is a prompting problem. It is the same problem engineering has always had: unclear requirements, no verifier, no feedback, no review. An agent just reaches the failure faster.
Which is the good news. The job moved. You no longer write the solution, you state it, and then you prove what came back. That is judgement you already have. This class is about where to put it.
One class, taught in modules
10 modules. Start anywhere. Most people start with the first one.
- See syllabus
Claude Code
The tool end to end: install it, drive it, and read what it is doing on your behalf.
- See syllabus
Codex App
The same work in OpenAI's Codex, and where the two tools diverge.
- See syllabus
Fundamental Techniques
The moves you reach for in every session, and why each one works.
- See syllabus
Advanced Techniques
What to do when a session stalls, and the patterns that only pay off once the fundamentals hold.
- See syllabus
Advanced Context Engineering
Give the agent what it needs to know, and keep it across sessions.
- See syllabus
Prompt Engineering
How to ask, so the answer does not need a second attempt.
- See syllabus
My Daily Workflows
The routines I run on production code, from the first prompt to the merge.
- See syllabus
Loopy AI
Loops that keep running after you close the laptop, and how much of the work you hand one.
- See syllabus
Creating Slack Agents
Put an agent in a Slack thread, so the work starts where the conversation already is.
- See syllabus
For Business
What changes when a business depends on what the agent ships.
Self-paced · lifetime access
Agentic Coding for Engineers
It starts at installing the tool and ends at the daily workflows I run on production code. Every update is included.
- Lessons
- Subtitle languages
- 10
What students say
The Agentic Coding School doesn't just show you what buttons to press, it builds genuine understanding from the ground up. By the time you're two-thirds through, you're not just using the tool, you're thinking in it.
SimeonStudent This is not the typical get-rich-quick hustle program you see on social media. Ray works deep in the inner workings of the models and knows how to be productive with them. I am convinced he uses Claude Code as well as or better than most Anthropic engineers.
Art SmalleyStudent I went from treating Claude Code like a chatbot to running it like a full development platform. The lessons on hooks, subagents, and CLAUDE.md files alone were worth it.
TylerStudent I already had experience with Claude Code, and this still took me further. Seeing your everyday usage throughout the course is genuinely valuable, not just some vibe coding app. I learned a lot from the multi-agent plan execution and the spec-dev command.
BernardStudent I am finding excellent value in the course. It serves as a vital reference library for me: whenever I hit a roadblock while pushing the tool's limits, I return to your material for practical tips. The biggest value is your experience as a builder.
DivyanshStudent This course is incredibly valuable for busy professionals who struggle to keep up with the fast-paced world of AI. He doesn't push unnecessary upsells but instead offers additional resources as genuine extras.
DeividasStudent
My newsletter
Stay in the loop
Every week I share what is actually working in my agentic coding setup. The Claude Code and Codex techniques, and the features worth your time.