Is the AI and Knowledge Job Market Lifeless?

information science was dying 7 months in the past?

It was additionally dying 2 years in the past. 

And dying 3 years in the past.

And to not point out it was additionally dying 5 years in the past.

Nonetheless, from the place I stand, that is positively not the case. Individuals nonetheless appear to land information scientist jobs.

I imply, I actually assist individuals do that each week in my teaching programme.

So, what on earth is happening?

Properly, on this article, I wish to break down:

• What the present information market seems like
• What it truly means to be an information scientist
• And, what you have to be doing to land a job on this present local weather

Let’s get into it!

Market Outlook

As a lot of you’ll know, there have been important layoffs throughout 2022 and 2023, with practically 90,000 tech staff being laid off in January 2023 alone.

In truth, it was so extreme that TechCrunch even created an archive of all of the layoffs that occurred throughout this era!

Nonetheless, in accordance with a research by 365datascience, information jobs weren’t that affected by these layoffs; they discovered that:

Apparently, our pattern’s largest group of laid-off staff didn’t maintain tech jobs — 27.8% labored in HR & Expertise Sourcing, whereas software program engineers got here in second with 22.1%. Advertising staff adopted them with 7.1%, customer support with 4.6%, PR, communications & technique with 4.4%, and so on.

For instance, solely 2.7% of individuals laid off from Amazon throughout this era had the title of knowledge scientist.

Based on one other research:

Knowledge science job postings grew 130% 12 months over 12 months after hitting all-time low in July 2023, whereas information analyst openings grew 63% in the identical time interval.

And we are able to additionally see that the wage of knowledge jobs as a complete has been rising through the years.

So, it’s clear that information science shouldn’t be dying by any means; if something, it’s rising.

Nonetheless, why does it really feel very arduous to get an information scientist job proper now, particularly on the entry and junior ranges?

To elucidate that, we have to look previous the numbers and actually perceive what the trendy information scientist is.

Knowledge Science Evolution

As an insider on this subject, let me inform you a secret.

Knowledge science shouldn’t be dying; it’s evolving.

10 years in the past, corporations would rent information scientists to tinker with machine studying fashions in Jupyter Notebooks.

In truth, that is precisely what my first information science job was like.

A knowledge scientist was like a Swiss Military Knife — one individual anticipated to do every thing from cleansing information to constructing fashions and presenting to the CEO.

Nonetheless, over time, corporations realised they have been getting no return on funding from this technique, in order that they grew to become extra stringent about roles and tasks to make sure they weren’t losing their cash.

This has led the information science job to turn out to be fragmented, and the title has turn out to be meaningless, as you’ll find information scientists doing fully totally different jobs at totally different corporations.

Basically, three flavours of knowledge scientists exist immediately.

Analyst

One of these information scientist is carefully aligned with the enterprise facet and primarily focuses on reporting workflows and experimentation.

For instance, you’ll:

• Get information from an organization database or different sources.
• Write some code that may be very linear and bespoke by nature, beginning with ingesting information, cleansing it a bit, then performing some EDA and a few inferential or fundamental modelling work.
• As soon as full, you set collectively a report that particulars the evaluation, gives visualisations and different metrics, and presents a advice based mostly on the evaluation’s objectives.

One of these information scientist is extra of an information analyst and usually requires extra enterprise area data.

Engineering

The main target of any such information scientist is on constructing and deploying options. This could be a vary of issues like:

• Inner software program tooling
• Machine studying fashions that drive resolution making
• Constructing libraries

This function leans extra towards software program engineering, however not like a software program engineer, it requires better data of maths, machine studying, and statistics.

These days, any such job has moved past the “information scientist” title and is now referred to as a machine studying engineer.

This isn’t entry stage place, and usually requires 2–3 years expertise in an adjoining function like a software program engineer or analyst first. So many graduates and other people with little expertise would battle to interrupt into this particular information science place.

Infrastructure

One of these information scientist is the rarest, primarily as a result of it has its personal title: information engineer.

The aim of this function is to construct the information infrastructure and pipelines to accommodate the enterprise’s information. This information is then used downstream by machine studying engineers, analysts and even non-technical stakeholders.

This function has turn out to be more and more necessary, particularly with the emergence of generative AI in recent times, which requires the power to successfully retailer massive quantities of knowledge and stream it with low latency.

At some corporations, you may additionally be an analytics engineer, which is a extra business-focused information engineer.

I do know, so many titles, its arduous to maintain up!

Junior vs Senior

A research revealed in September 2025 has been making fairly a number of waves within the information and machine studying house.

The research examined 285,000 corporations between 2015 and 2025 and the way their adoption of GenAI has affected their hiring processes for junior and senior positions.

Observe: this is applicable not simply to information scientist jobs however to all jobs at these corporations.

You possibly can see within the plot beneath that hiring for senior positions remains to be rising, whereas hiring for junior positions is lowering.

This makes intuitive sense, as juniors’ tasks are probably simpler to automate with AI than seniors’ because of the wealth of expertise they’ve constructed through the years.

What I wish to clarify, although, is that corporations aren’t making juniors redundant nor are there no extra junior positions left in the marketplace. 

Most individuals will take a look at this graph and suppose that the junior information science market is changing into extinct. However that’s objectively not the case.

Hiring remains to be taking place, however the fee of latest positions being posted shouldn’t be rising. The provision curve stays unchanged whereas demand stays excessive. 

That’s why it feels so arduous to get an entry-level job these days.

What Can You Do?

I’m going to be sincere, it’s changing into extra aggressive to interrupt into information science, but it surely’s not unattainable.

Gone are the times when all you wanted was fundamental Python and SQL, and having accomplished Andrew Ng’s Machine Studying course.

These are issues everybody has these days, so that you must go the additional mile and differentiate your self greater than you used to.

There are lots of methods of doing this, for instance, you undertake and specialize in sure technical domains like:

• GenAI
• Mannequin deployment
• Time collection forecasting
• Suggestion techniques
• Area-specific experience

Specialists are arguably changing into extra necessary as data is more and more democratised by AI. Having deep experience is sort of a rarity these days.

An alternative choice is to go for a lower-level place, like a enterprise or information analyst function, that’s extra pleasant to junior and entry-level positions, after which slowly construct your method as much as a full-time information scientist place.

You must also give attention to areas that AI can’t actually change:

• Speaking successfully with totally different audiences
• Understanding the enterprise impression of your work
• Essential pondering and realizing what drawback to unravel
• Sturdy fundamentals in maths and statistics
• Relationships and community

These are timeless abilities, particularly the final one.

You may need heard the saying:

It’s not what you recognize, however who you know

I truly disagree with this.

The actual energy is in who is aware of you.

When you have a stable community and relationship with many individuals within the subject who worth and belief you, you’ll be able to faucet into this to get referrals, alternatives, and even develop your community additional.

The leverage this gives is unbelievable. I at all times inform my teaching purchasers that referrals and networks are actually the golden ticket to getting top-end information science jobs.

And all it requires, is simply effort and pushing your self out of your consolation zone to talk to individuals you wish to join with.

Applied sciences will come and go, however precise human relationships will stay central on your entire profession.

The reality is, you’ll must reinvent your self each 3–5 years as an information scientist, since know-how shifts in a short time.

So asking “Is information science dying?” misses the purpose.

Knowledge science is at all times technically dying because it’s persistently evolving and reworking.

However that’s what makes it thrilling.

And if you’re prepared to up-skill and put in additional effort than others, you’ll be rewarded very properly.

Should you’re able to dive into information science after studying this, that’s an important first step. 

However right here’s the truth: I’ve been on this subject for 5 years, and looking out again, I spent my total first 12 months on duties that have been a whole waste of time. In immediately’s hyper-competitive market, you don’t have the posh of trial and error.

To keep away from my errors and speed-run your progress, take a look at this information the place I map out precisely how I might turn out to be an information scientist once more.