Expertise in the Age of AI
Brian Kihoon Lee
Essays
Expertise in the Age of AI
2026-05-12
Tagged: llms
Does it make sense to hire junior engineers in the age of coding<br>agents?
Junior engineers are expensive, both in salary and seniors engineers’<br>time. This cost was partially recouped through code contributions, but<br>today, it’s more effective to directly maximize the output of your<br>senior engineers. The hiring market reflects this trend: senior<br>engineers have an easy time finding jobs, while fresh CS grads are<br>having their worst years ever. And yet, OpenAI, Anthropic, and many top<br>companies continue to compete fiercely for junior talent. What’s going<br>on?
In this essay, I’ll explore the changing nature of expertise in the<br>age of AI.
Math as an analogy
I think it helps to think about the impact of AI in terms of math,<br>which had its AI moment half a century ago.
There used to be a job called “calculator”, which was a human who<br>could do math calculations accurately and quickly. These people balanced<br>books, calculated artillery firing angles based on distance and wind<br>adjustments, calculated optimal hull shapes for ships and aircraft<br>bodies, and so on. This job doesn’t exist anymore, and the last serious<br>use of abaci and slide rules was in the 1970s, due to the invention of<br>the scientific calculator. Calculators have only become more<br>sophisticated over time, with today’s numerical modeling software<br>running full scale physics and engineering simulations. (For the purpose<br>of this essay, I’ll use “calculator” to mean everything from basic<br>calculators to modeling software.)
Despite the existence of calculators, we teach and expect people to<br>learn algebra, geometry, and calculus in high school. Continuing into<br>the college level, we expect STEM majors to learn multivariable<br>calculus, ODEs, PDEs, statistics, and linear algebra. Upon graduation,<br>the vast majority of them use calculators every day and forget how to do<br>all but the most basic mental math.
There are two basic explanations for this discrepancy:
(Signaling hypothesis) The STEM degree filters the set of people who<br>can both learn and persist through four years of difficult math.
(Skills hypothesis) Struggling through math classes imparts some<br>hard-to-quantify mathematical intuition that is valuable for operating<br>today’s calculators.
As a formerly strong believer of the signaling hypothesis, I am now<br>increasingly buying the skills hypothesis (let’s say ~50% attribution to<br>each cause). It’s clear that senior engineers today are far more capable<br>of using coding agents than their junior counterparts, and a large<br>portion of this is due to having struggled through 5+ years of writing<br>code manually.
A job market in flux
Currently, the level of computing intuition needed to additively<br>prompt the coding agents sits at roughly 5 years’ experience level.<br>Today’s seniors were lucky enough to get paid to build their computing<br>intuition, but the gap grows as coding agents continue to improve.
In between coding agent improvements and natural variation in<br>learning aptitude, maybe 50% of new CS graduates will not be able to<br>catch up, ever. Some senior engineers will also eventually fall behind<br>the curve despite their head start.
To answer the opening question of the essay: only some junior<br>engineers are worth hiring, specifically, the ones who are good enough<br>to reach some useful threshold of “coding intuition” within ~2-3 years<br>of having graduated. Since there are not very many of these graduates, a<br>small number of elite companies compete fiercely for this talent.
The second-class tier of software consultants will continue growing,<br>expanding the total size of the job market, but I don’t anticipate that<br>their salaries will grow anywhere close to as rapidly as today’s senior<br>engineers.
Everyone should learn some<br>coding
Even as the bar to get into software engineering rises, I still think<br>everyone should learn some coding. Too often, I see people treat<br>computers as appliances - capable of doing what they were built to do,<br>but nothing more. If you don’t think of computers as scriptable or<br>programmable, then you won’t ever think to ask AI to automate something<br>for you! The same is also true for many other fields, too! Math, law,<br>taxes, medicine, DIY home repair, etc… Abundant and cheap expertise is<br>now available for just $20/month, if only you know how to ask.
I would say that the major unlocks are at:
1-2 weeks: Basic understanding of what the field is about and what<br>general words to use when asking the AI to do something.
1-2 months: Basic understanding of how and when to ask the AI<br>something.
4-6 months: Ability to check the output for correctness (using<br>external sources as needed).
If you’re already a software engineer, you might consider dabbling in<br>data science, frontend, backend, security, and performance<br>optimization/profiling – all of which are distinct skillsets.
Here’s a data science example of a “how + when + correctness”: A<br>coworker was...