AI-Driven Development
course

When we started, we didn’t fully understand what the course would be like, but by the end I realized how well the assignments are split: each one doesn’t repeat the previous one – it covers a separate tool for working with AI. This was very valuable for me: on my project I had just started actively bringing AI processes into my team’s development, so I applied what I learned in practice right away. I liked how fundamental and structured the theory was, the very good feedback on homework, and how Sergey showed everything both with examples and on his own work projects.

I started with chat mode and an IDE plugin, and now I’m comfortable working in the console, I’ve written a couple of my own MCP servers and moved to a Pro subscription because I was burning through my limits. The course gave me two key skills: running several projects in parallel with agents while keeping several contexts in my head, and seeing AI in development as an engineering practice rather than just a code generation tool. I especially value the seventh homework: entering a new project with an agent will, in my opinion, become the new standard for interviews.

I came to the course disliking AI tools: unreliable, unpredictable, non-deterministic. But the very first assignment showed me the difference between what and how you give the model as input and what you get as output. I saw my own influence on the result.

Over the course I came to see that programming with and without AI tools are two different skills, and working with AI is closer to management: you try to guess how your task will be misunderstood, where to put checkpoints, how to break it into parts and what you’re willing to risk.

During the course I tried AI in development on a real work task for the first time: a plan, step by step, a check after each step. I started from zero and felt how much the course helped – without it I would have done everything differently. Thank you very much!

I realized you should learn from enthusiasts, and Sergey and Pavel are exactly that: they develop the topic in a very unstable environment where everything changes often, and they know how to present the information in a generalized way and teach you to work with it.

Before the course I worked with agents intensively through special plugins and thought it was cool – it turned out to be not nearly as cool as I thought. After the course I switched to the console, and it truly changed my paradigm: now for any task, not necessarily a programming one, I open a new worktree, brainstorm with the agent, make a plan and keep the results as reusable groundwork. It’s a significant qualitative shift that came precisely after the course.

The course is very well split into topics, each of them matters, and you can apply all of it right now. The reviews of my sessions with the agent were very useful: the mistakes listed at the beginning of the report and recommendations on how to talk to the agent when it gets stuck in a loop.

I liked the course – it broadened my horizons: now I know where to look if I need to go deeper. There has already been a lot of feedback about the homework. I also found the homework hard to read, so I used an assistant to do it. Because of that, in places it came at the expense of understanding the homework itself. I also liked it when Sergey showed how he uses AI in his own everyday tasks.

This is my first course, which I decided to take on a friend’s recommendation. Before it, I took someone else’s lecture course on system design and thought the AI course would be roughly similar in the amount of theory and assignments. The volume turned out to be an order of magnitude higher (which made me very happy) – so much so that, with my job, the main course time wasn’t enough, and I’m going to go through all the material again on my own.

I liked that the course immerses you in modern and future development. I’m a QA engineer, and it was tough, but it’s a great experience – seeing how the development process will work now and how you can influence it at an early stage. I’ll also add that you need to set aside more time than the course recommends: there’s a lot of material, and it isn’t easy. The feedback is very detailed and to the point.

I wasn’t a beginner: I already knew some of this and used it actively before the course. But the course pushed me to try things I would never have tried on my own – for example, working with MCP as a creator rather than a user. That’s useful: at work we build services that expose their own tools to agents, and now I understand how they work under the hood. The way the homework is structured teaches you to work with an agent effectively – to manage context and get results with weaker models. The n8n integration in the sixth homework showed me which design decisions I had got wrong before – and taught me to design the system differently. The most important thing for me was the planning assignment: a plan, concrete steps with a checklist, a check after each step. I applied this approach to refactoring old legacy code at work, and it helps me get complex tasks done faster, while the plan itself gives the agent a history of what has already been done.

Before the first course, AI-Driven Development, I had almost no experience working with agents. The course gave me an understanding of what kinds of agents there are, which ones to use and what tasks you can solve with them. And something I didn’t expect at all: you can build products on all of this, and do it yourself, almost single-handedly.

The second course, Product AI Development, gave me more confidence: you can build fairly robust production products on generative AI, and it’s closer and simpler than it seems at first glance. I had experience building various systems, but I hadn’t built systems like these before – now I do. The topics are comprehensive, the homework is substantial, I had to dig into it, and I liked that. Most of all I liked the review of my AI work sessions: seeing from the outside, with an assessment, how I use AI is a very good tool, and I’ll keep using it.

For an entry level, the list of tools and everything about context is exactly what matters to understand in a company of any scale today.

I'd especially highlight assignment M7: it captures what many are actually trying to do in production right now – moving to local models for personal data and cloud models for everything else. I did all the assignments with pleasure.

The course gave me a great deal in terms of theory and broadening my horizons in LLMs – before it, I had barely moved in this direction. I saw the announcement of the first version of the course in the student chat and, without a second thought, signed up right away: it resonated with me.

The practice, considering this was the first version, I really liked – the assignments were interesting, especially the ones using n8n (the fifth and seventh homeworks).

And the most valuable thing for me personally: a good course doesn't just make you a specialist, it gives you motivation. This course genuinely got me interested – I'm passionate about it and want to keep growing in this direction.

Before the course, I knew only the very basics about working with LLMs. And here a whole universe seemed to open up for me – so much new that you can't help wanting to dive in.

I'll definitely keep coming back to the materials we covered, rewatching and repeating things with my own hands.

I genuinely enjoyed it, and it was comfortable being on the course and in the communication. The start is made – and I'm glad we may see each other at the next stage.

I want to express my gratitude to Sergey. I'm a seasoned techie – and that makes me admire the topics he raises all the more; at times it even makes me feel a little ashamed in a good way, and that motivates me to push further, because you see: it's possible.

My main goal was to broaden my worldview in this new technology – and in my opinion the course did that 100%, covering and touching on every aspect across all directions.

In learning, responsibility is split 50/50 between teacher and student – and the teacher delivered his 50% to the fullest. The rest of the baton is on us, and now it's up to us to decide where to go deeper.

Before the course I used AI more as an assistant for work tasks, without building a production pipeline. I was a skeptic and always kept the final decision to myself. The course, though, genuinely broadens your horizons: a lot of information and a much deeper understanding of where to dig next.

I started applying part of what I learned right away in my work, on the research side – gathering and analyzing information. These were my own personal experiments with how AI can be used for academic purposes.

I now work as a programming teacher at a coding bootcamp, so I have something to compare against: what this course delivers is an order of magnitude above what many European bootcamps offer. I was pleasantly surprised by the level. For me it was a very useful course precisely in terms of broadening my horizons and understanding how to work with agents going forward.