Show HN: I analyzed 1500+ job ads to find the most wanted skills by recruiters
skillsets.tech
skillsets.tech
This might seem nitpicky, until you’ve had a recruiter ask how skilled you were in JIRA and demands you tell them a story about a time when you used advanced JIRA skills to solve a problem. It becomes a checklist of things that really don’t matter that much compared to actual skills.
Employers, this is also why you can’t find good candidates. They might not have a lot of “skills” in the way they’re being defined.
The recruiter was tasked to find candidates with a recommendation system background but the only way they know to do that is look for that exact word.
It's like saying my skills in "English" are equivalent to Maya Angelou's skills in "English"
Skills are things can do regardless of the technology/language. My skills are in computational math. I could write you a good algorithm in JavaScript with maybe a day or two of ramp-up/review of the nuances of the language. I could not (quickly) write a (good) web application in the same language.
I can deploy complex distributed computational code using EC2, I cannot even begin to understand the IAM. Do I have "Skills" in AWS?
It's such a pain when I'm reviewing resumes. I don't care about the language, I care about what you can do, and that's so hard with the checkbox system we've devolved into.
Much of my career has been based around low-level C and bit-twiddling optimizations for micro optimizations of math for speed and floating point accuracy. My current job is at a Java shop. When interviewing for that work, I hadn't touched Java since CS101 in college. I was lucky that I was interviewing at a small company that didn't have recruiters, but had me talk directly to the team and they could understand that the skill was "Math" not "Java". I still can't understand how Gradle or the rest of the Java build system we uses works, so it's good we have someone with "Skills" in DevOps to do that.
I hope the industry recognizes this soon - I think you're right, it's probably the biggest impediment to finding good candidates.
But if you need them, the business is likely f'ed and you should leave. Rapidly.
1500+ jobs in which location? Which country? Which job titles did you search for?
By the way, 1500+ is a start but insufficient to determine trends. For instance, 48,344 SW jobs were posted in Jan 2025 alone [0].
It is a nice project to learn and show potential employers. Well done.
I’ll definitely add more context and details to improve clarity in the project. I agree that 1500+ job ads are a starting point rather than a definitive trend analysis. That’s why I plan to continue collecting and analyzing job postings over time to get a more representative dataset.
To answer your questions — the job ads were primarily sourced using search queries tailored to each category. For example, the "Frontend" category included searches for "Frontend Developer" / "Frontend Engineer" / "Frontend Web Developer". Similar criteria was used for other categories too.
The dataset covers multiple countries but is predominantly focused on regions: North America and EU.
Only a single mention of C++, and yet, it represents about 10% of GitHub, about the same level as JavaScript. But JavaScript gets 375 mentions.
I also don't see a single skill that relates to embedded software, except C, but C is not just for embedded software. And thinking about it, I don't see much about system programming either (no Unix/Linux?). And nothing about 3D graphics either (OpenGL, engines, ...).
Guess I'll be homeless if my current job doesn't work out??
After recently graduating from college, I set myself a personal project to develop a simple tool for analyzing job advertisements and extracting the most in-demand skills, technologies, and tools for various IT positions.
By automatically scanning and processing data from more than 1500 job ads, the project identifies key skills and technologies that employers and recruiters are looking for.
You can browse various IT positions to explore the skills needed for each, including Frontend, Backend, DevOps, AI/ML, Data (Engineering/Science), Full Stack, UX/UI, Technical Writing, Quality Assurance, iOS/Android, Management, and Cybersecurity.
The project was already announced via Reddit, receiving over 150,000+ views on the launch post!
Average Trend – This is calculated by taking the percentage change between consecutive months and averaging these values. It gives a general sense of how skill mentions increasing or decreasing over time.
Median Trend – The percentage changes between months are sorted, and the median value is taken. This reduces the impact of extreme fluctuations and provides a more stable measure of growth.
CAGR (Compound Annual Growth Rate) Trend – This is calculated by comparing the first and last month's skill mentions count, then applying the CAGR formula to estimate a steady growth rate over the observed period. This measures long-term trends rather than month-to-month volatility.
If there are fewer than three months of data, trends are marked as "N/A" due to insufficient data for meaningful analysis.
Something is not right with the data. This is a fundamental.
* 99% of jobs advertised online are fake, including HN's Who's Hiring. They are posted just because some manager wanted to fake growth, wanted to pretend fairness for a position that will be given to patronage, forgot to remove an add posted long ago, etc. Like in online dating, online job postings is mostly a scam.
* The "most wanted" skills are somewhere between the "junior skills that are necessary but not sufficient" and "rat race". They are the skills most likely to be replaced by AI or cheaper workers from 3rd world countries.
I've applied for online jobs, I've been through the pain.
There are only 3 ways of applying to online jobs that preserve your emotional health:
* Use an automated tool like loopcv, Sonara, Jobscan, LazyApply, SimplifyJobs, Massive, etc. Yes, they're crappy. But with enough quantity, quality of application becomes irrelevant.
* let recruiters find you, don't search for them
* look for positions in fora dedicated to specific/niche technologies (e.g.: Reddit, Dischord, etc).
Second part is mostly true, they might be, but what's exceptional talent if you can train anyone?
Doomer much?
1. Node.js
2. React
3. JavaScript
4. TypeScript
5. Python
6. Java
Would completely avoid super specializing in these 'skills' unless you are the top 1% in them as these employers are extremely selective and expect candidates to be in the top 1% out of thousands of applicants.
But any self-respecting JS full stack dev would want to have a good level of familiarity with the first 4, so some level of experience with them is absolutely a +++
But I have no counter-example to give because LinkedIn is abjectly useless, shows me 192,518 jobs for "c++ in European Union", where the first ones in the list title "Sales Engineer".
Really, the only reason I'm still on LinkedIn is because everyone is there but boy would I leave the motherfuckers if a valid alternative would show up. It's abysmally bad and I hate it deeply, profoundly and sincerely and I really hope it (LinkedIn) dies in flames and Microsoft along with it.
Frankly: imo if you need a website/tool like this/find this website helpful, you might consider it a sign you need to start thinking about life more generally. :)
I can't think that producing a new abstraction of a signal that was already readily available in a way that is less information rich not more, is a good way to be competitive in the job market. Basically: it's lazy and encourages others to be lazy, and it's lazy in the worst kind of way (lazy that seems it isn't).