A New Chapter in My AI Journey

Career transition, first impressions of video content, and a sneak peek at my new course.

Sep 20, 2025

Welcome to Neural Bits. Each week, I send one article on practical, production-ready AI/ML Engineering to help you learn and upskill in your AI/ML Journey. Subscribe and join over 6000 engineers who learn how to build real-world AI Systems.

During the past two weeks, a lot has happened.

I’ve completed the code development and video recordings for the Kubrick Open Source Course (YouTube), which I developed alongside

Miguel Otero Pedrido

(from The Neural Maze).

This Monday marked my last day at Everseen, where I spent the past four years training models, building Vision AI systems, and running MLOps workflows.

I’ve stepped into a new role as a Senior AI/ML Engineer, where I’ll be working on large-scale AI (GenAI) systems, something I’m genuinely excited about. I’m already diving into the codebase, workflows, and getting up to speed with projects and documentation.

That means, even more production-ready AI Systems insights to share with you here.

My schedule has taken a hit with the transition, but this post isn’t about taking a break; it’s about the next steps I’ve been planning for a while.

If you’ve been following my newsletter, you know I’m focused on production-ready AI and the components of real AI systems. These are the areas I think not too many talk about, and I want to continue filling that gap.

So in this article, I want to touch on 3 things:

What I’ve learned, transitioning between AI roles
My Video Content First Impressions
[NEW] Course on Production-Ready AI Systems

Let’s get started!

Two thoughts when transitioning between roles

Leaving a role after years in it is never easy. It’s not just the projects or features you’ve worked on and shipped, but more about the failed demos, inside jokes on calls, `git blame` pranks on your teammate’s PR, and real friendships you’ve built along the way.

Moving forward, I wanted to outline three things that I think are key for everyone:

Keep your inbox open
Even if you leave, make it easy for colleagues to reach you, and offer yourself to help when your help is needed, considering your time. Be open to short chats or advice on things you’ve worked on.
Leave it better than you’ve found it
And by this I mean, clearing up the backlog, documenting everything, and doing your best to ensure that whoever takes on from where you’ve left, will ramp up quickly and thank you for that. This is a treat that every Engineer should aim for.
Don’t be like Joe

Comfort Zone
After years in a role, it’s easy to fall into comfort zones: how you organize your work, how you communicate with teammates, or how you approach problem-solving can all become routines. A new environment shakes that up. I think this is beneficial.

Video Content - First Impressions

You might’ve seen that I’ve uploaded a few videos in my previous posts.

That’s something I was planning to do for a while, but didn’t quite manage to get the time for. The course I built with Miguel was the kickstart for that, as I recorded over 1.5 hours of video code walkthrough, going over key technical components.

Here is me explaining how React works, for Data Scientists and AI folks.

Why React?
Some AI tooling is increasingly moving to JavaScript, with tools like Transformers JS, n8n, MCP, and Mastra. Learning a frontend framework is becoming a valuable skill for AI engineers.

I like the video format, as it allows me to show and explain, in real-time, what I’m talking about. You could expect to see me more often, posting Videos, going through complex AI concepts, and live-coding from now on.

Help me shape and structure video topics better:

Also, please leave your thoughts in the comments on what else I could do to make these videos better 👇 Thank you!

A New Course: Production-Ready AI Systems

I’ll keep this short, as I’ll be rolling out big updates in the upcoming weeks.

I’ve been working on a new project that combines Vision AI, Generative AI, and Agents to build a fully-fledged AI system from the ground up. This will be my most advanced project yet, and I believe the only one that covers advanced AI concepts and key AI libraries and frameworks at a low level.

🥁🥁 Here’s a sneak peek!

We’ll build an Edge Multi-Agent Vision System for Wildlife Conservation

What’s inside?

Building End-to-End MLOps Pipelines
Finetuning & Evaluation
Model Optimization - advanced model optimization techniques.
Perception & Multimodal AI - we’ll classify actions from videos
MCP Servers & Agent-to-Agent - we’ll build a network of AI Agents
AI System Design and a lot more

To properly organize this, I’d love your input! Please take 1 minute to answer these 3 polls. Your feedback will help me structure the course modules in the best way possible.

Have extra details to add? Please leave them in the comments.

What to do Next

I’ll keep you updated with everything you need to know, but for now, stay tuned for the next articles and videos I’m working on.

This next project is quite complex; I’m not sugarcoating it, but I want to make sure I ease the way towards understanding every piece of the puzzle.

My goal is to make sure you understand how everything fits together.

Thanks for reading! Follow this newsletter on YouTube as well, as I’ll be posting longer videos to prepare you for some of the hands-on concepts we’ll be covering in this new project.

👋 Cheers