Hello Reader,

Data science jobs are evolving.

In my 10 years working in data, I’ve never seen a shift this massive.

🤯 Expectations have changed:

A recent study by Lightcast found that from May 2024 to May 2025, Data Scientist was the job title that most often referenced Generative AI (3,301 unique postings).

Some of these postings talk about AI literacy - collaborating with AI tools, using copilots, experimenting with LLMs.

But most of them are about building and deploying AI-driven systems.

We’re talking RAG pipelines, AI agents, prompt engineering, vector databases, etc.

Even DevOps skills like CI/CD and cloud deployment.

A few years ago, this wasn’t expected from a typical data scientist. Heck, GPT wasn’t even on most people’s radar.

Now it is - particularly for more senior data scientists on the team.

Many of my data scientist friends have been pulled into GenAI projects at work, without necessarily changing their job titles.

They’re integrating LLMs into internal tools, or building prototypes for business teams.

In other words, they’re doing parts of an AI engineer’s job.

If you think about it, it actually makes sense.

As a data scientist, you sit at the intersection of business, data, and engineering. You understand the metrics. You talk to stakeholders. You can code well enough to build things.

That puts you in a strong position to identify AI use cases and turn them into working prototypes.

So building with AI is becoming less of a “nice-to-have” skill and more of an extension of the standard data science toolkit.

Even if the job title hasn’t changed, the expectations have.

🧐 Another important shift:

The bread-and-butter skills of data science, e.g. exploratory data analysis, visualization, extracting insights, prototyping statistical models,... can now be automated or dramatically accelerated with AI.

For me personally, there’s always been something deeply satisfying about exploring a messy dataset and uncovering patterns. It feels creative, it feels like you're a data detective.

Now that AI can do a big chunk of that work in much less them, it forces you to rethink where your value really lies.

Whether we like it or not, data science day-to-day work is changing. That means there’s a lot to learn, and some things to unlearn.

👉 How to adapt:

I don’t want to repeat the usual “learn AI skills” mantra and add to the “I’m behind with AI” feeling a lot of us feel.

Although they are important right now, I still think the more important skills - the ones that make you resilient in the long term - are the ones AI can’t easily replace:

From what I can tell, AI can’t sit in a messy meeting, read the room, balance competing priorities, and own the outcome.

That’s still human work 😊.

Have a great week ahead!
Thu

P.S: Work with me:

If you want a comprehensive course from Python fundamentals to building AI applications, check out my Python for AI Projects course. It’s packed with everything you need to build solid fundamentals and transform your skills in 2026.

​Learn more about the ​course.​