“I want to learn AI” – This is what a lot of people interested in Le Wagon programs start their advisor call with.
And it is a legitimate wish. The acronym AI itself is everywhere, used as a token of up-to-dateness and promising to deliver wonderful results, or tagged on web applications in the hopes of raising an extra hundred thousand dollars.
But… as a top tech education provider, we need to address this critical question: what does “learning AI” actually mean? And more importantly, what objectives will it help you achieve?
At Le Wagon, we believe that each learning step helps you achieve specific goals – whether this goal is only broad understanding, staying up to date with the latest tech trends, launching your own company or steering your career towards a tech role.
So let’s dive into the different aspects of “learning AI”, what goals can they help you achieve, and how we can help with these goals!
While the word AI became popular thanks to chatbots (I am looking at you, ChatGPT) and Large Language Models (LLMs), the field obviously did not originate with them. With that in mind, one step in an AI learning journey is to understand the different flavors of AI.
This step is about AI literacy: grasping key concepts without needing to code. You will learn what machine learning is, how neural networks try to mimic the brain, how generative AI can produce realistic text or images, and also what ethical challenges may come with it. More critically, you will also learn that AI did not start with ChatGPT, and that there are hundreds of other applications besides text generation (impressive, right?).
Interestingly, this starting point is also strongly connected to data-related topics. Why is access to data so important when it comes to training Machine Learning models? How is that data collected or generated? And why does low-quality data almost always lead to low-quality products?
Gaining an overall understanding of the field of AI is a critical step which will help you figure out a learning path. The primary goal will hence be to “unpack” AI: grasp the nuances, use more specific vocabulary, and understand the limits.
As a working professional, in tech or not, you will also be able to plan around it, and make better choices for your career and your organization.
For most people, “learning AI” means learning how to use AI-powered tools effectively.
And curiosity and a willingness to experiment will be more critical traits than coding skills. Instead of building algorithms, the goal is to leverage chatbots and automation (sometimes, but not always, AI-powered) to get things done faster, smarter, or more creatively.
Some time ago, learning how to use AI basically meant getting the most out of chatbots: having a clear framework for how to prompt them, understanding the critical idea of context and its limitations, staying alert to potential hallucinations.
But tools have switched to second gear, and do not stop at telling you things anymore. They can now do things for you: read and autonomously answer your emails, organize meetings for you, follow non-deterministic workflows… all the way to taking full control of your computer (a prime example being Claude Cowork). Welcome to the world of AI Agents.
The goal here is often up-skilling: without having to become a developer, learning how to use AI will help you become more productive, creative or data-savvy in your existing role.
Most people starting up with AI-powered tools can easily imagine how it can help automate their current daily tasks. But once you understand how to truly use them, you will realize that entire aspects of your job can be reinvented.
One of the most convincing and talked about applications of large language models (LLMs) is their ability to produce reliable code. In essence, code is a very structured and deterministic “language”, which is something LLMs excel at.
For that reason, it is becoming increasingly easy to “build with AI”: rolling out entire applications (web or mobile) in just a few hours, using tools like Claude Code, Lovable or Replit. This works really well when what you need to build is a small scale application, often starting from scratch. And once again, very often, you don’t even need to know how to code.
When dealing with larger applications, SaaS or enterprise software, producing code is only one step in the development process: individuals and teams need to plan, collaborate, and decide what needs to be built based on usage and customer feedback.
With these tools, the role of a developer is also evolving. We’ve been interviewing CTOs, tech leads and engineering managers over the past few months, and one theme keeps coming back:
“I am looking for engineers with a product mindset – people who can ship code, have solid technical foundations, but who also have a strong user-centric approach and understand the subtlety of building a product”.
From the solo entrepreneur to the senior architect, a new set of skills is appearing: building with AI.
If you are a solo founder, building with AI is transforming the product testing phase: no more spreadsheets or landing pages, you can now build the application of your dreams in a few days.
And for some people, building with AI will be the first step to a bigger journey in tech, as a developer or product manager.
Typing a prompt in Claude and watching it generate an answer is the end result of years of work, and LLMs themselves are only the tip of the “AI” iceberg.
Hundreds of companies are training niche Machine Learning models, using data specific to a goal (Te Hiku Media is a great example, a company that built a Māori Speech AI Model). “Building AI” means going down this rabbit hole to discover all the roles that make today’s Artificial Intelligence, from AI Research Scientist to Data Engineer.
At a broader societal scale, building AI is also connected to alignment, safety and ethics. Research roles related to this means asking tough questions, for example: “Do AI-powered tools behave the way we, as humans, want them to?”.
We won’t lie – having a solid background in programming will go a long way. But potential implications are so broad that various STEM backgrounds (such as geneticist or neuroscientist), as well as non-STEM like linguistics or philosophy backgrounds will have a critical role to play building AI.
If you are already working as a web engineer, this can be your next step: broadening your role to integrate more data-related topics, potentially turning yourself into a “full-stack AI engineer”.
In various fields, researchers now have to master a foundational programming language (usually Python), and also to test Machine Learning applications in their area of expertise.
At Le Wagon, we’ve developed a full suite of programs geared to help both individuals and companies accelerate their AI learning, whether they want to “just” understand it or help build the AI usage of tomorrow.
Our brand new AI Product Builder and AI & Data automation courses now offer a shorter 40-hour format. These courses keep Le Wagon’s unique pedagogical approach: access to videos and exercises through our content platform, a buddy system that promotes pair learning, and a learn-by-doing style focused on building actual projects.
From our original Web Applications Development or Data Science intensive bootcamps to our shorter Skill Courses and B2B offerings, our programs are designed to address the needs of both tech and non-tech professionals.
Use the four parts above as a map, a starting point, and ask yourself a simple question: What do I really want to achieve with AI?
In today’s digital age, web development plays a crucial role in shaping our online experiences.
After working as a Financial Analyst in Japan, Daniel uprooted his life and moved to
We celebrated our third anniversary a few months ago, and we thought it would be
Le Wagon is up amongst the best things that I’ve ever done. I feel like
The hiring day takes place at the end of each bootcamp, where recruiters from leading
For several years, Tiffany dealt with tech companies and employers as a recruiter. She realised
