I found myself wondering recently:
Are job descriptions asking less about languages than they used to?
So I went looking.
I started going through postings to see if it was real…
And the pattern showed up pretty quickly.
Job descriptions aren't asking for languages as much.
Instead, I'm seeing more emphasis on:
- specific AI tools
- platforms
- workflows
Less:
"strong experience in X language"
More:
"experience with [tool] or [AI-assisted development stack]"
That's a shift. And I don't think it's accidental.
AI has lowered the barrier to writing code. You don't need to know everything about a language to get something working anymore.
That's powerful.
But it also raises some questions.
What Are We Optimizing For?
If language expertise becomes less important, what replaces it?
Because something always does.
Right now, it looks like we're replacing:
deep understanding of a language
with:
familiarity with tools that generate or assist with code
That's not inherently bad.
But it changes how decisions get made.
Concern #1: Choosing the Right Tool
If fewer people have deep language experience, then fewer people are equipped to evaluate:
- trade-offs between languages
- long-term maintainability
- performance implications
The risk isn't that we pick a bad language.
It's that we stop asking:
"Is this the right language at all?"
Concern #2: AI Bias Becomes Industry Bias
AI tools don't treat all languages equally.
Some are better supported. Better trained. More represented.
I ran into this firsthand recently.
I was exploring a project idea and decided to use AI as a sounding board. I'd already done most of the thinking—worked through trade-offs, had a direction in mind.
But as I talked through the design, it kept nudging me.
Away from what I had chosen…
Toward C#, Python, or Go.
Not once. Repeatedly.
So afterward, I ran a small experiment. I started having similar conversations—same kinds of problems—but across different languages and frameworks.
And it didn't take long to notice a pattern.
Some languages were consistently favored. Others were… gently discouraged.
Not explicitly.
But enough that, if you weren't paying attention, you'd end up somewhere you didn't choose.
Over time, that can create a feedback loop:
- AI favors certain languages
- developers use those languages more
- those languages become even more dominant
Not because they're always the best choice.
But because they're the easiest.
And languages that fall outside that loop?
They start to fade.
Concern #3: The Perception of Skill Changes
If AI can generate working code...
then the perception becomes:
"How hard can this really be?"
That's not new.
But it gets amplified.
Because now, more people can produce something that looks like a solution.
The difference between:
- working code
- good code
becomes harder to see.
I've seen how that perception can play out.
I worked at a place where the owner of the company was heard—more than once—saying:
"I don't understand why I have to pay engineers more than data entry people. It's the exact same thing."
That wasn't coming from malice.
It was coming from misunderstanding.
A lack of visibility into what actually goes into engineering:
- the decisions
- the trade-offs
- the long-term consequences
AI doesn't create that misunderstanding.
But it can reinforce it.
Because when code becomes easier to produce, it's even harder from the outside to see the difference between:
- something that works
- and something that's built well
And when that distinction disappears…
so does the perceived value.
Concern #4: A Tighter Job Market
Lowering the barrier to entry does something else. It increases the number of people who can compete.
That's great for access.
But it also means:
- more applicants per role
- more noise in the hiring process
- harder differentiation
Especially for engineers early in their careers.
This isn't all downside.
There Are Real Advantages
1. Easier Transitions Between Languages
Switching stacks used to be a real hurdle.
Now?
It's easier to ramp.
That opens doors for engineers who want to:
- explore different ecosystems
- move into new roles
- avoid being locked into a single stack
2. Access to New Industries
Language requirements have historically acted as gatekeepers.
If you didn't have the "right" experience, certain roles were just off-limits.
That's changing.
More people can move into:
- industries they care about
- problems they're interested in solving
That's a good thing.
3. Working Across Environments Gets Easier
AI reduces friction when jumping between systems.
You don't have to context-switch as hard.
You don't have to remember every detail.
That makes engineers more flexible.
So Where Does That Leave Us?
I don't think language knowledge is going away.
But I do think it's being deprioritized.
And that means we need to be more intentional about what we value. Because if we're not careful, we don't just lower the barrier to entry.
We lower the standard.
The real skill has never been writing code.
It's understanding what should be written…
and why.
The best engineers I've worked with know multiple languages.
Not because the languages matter that much—
but because thinking in different systems does.
They can look at a problem from every angle:
- performance
- maintainability
- trade-offs
- long-term impact
They don't just solve the problem in front of them.
They understand the shape of it.
That's the skill.
And no tool replaces that.
At least not yet.