Everyone's banging on about AI talent shortages. Every recruiter's LinkedIn is clogged with posts about how impossible it is to hire ML engineers or prompt engineers or whatever the flavour of the month is. But while we're all fighting over the same pool of ChatGPT specialists, there's a much more fundamental crisis brewing.
And it's happening right under our noses.
The backend engineering drought
I've spent the last six months trying to fill backend engineering roles for three different SaaS clients. Not cutting-edge stuff. Not bleeding-edge ML. Just solid, experienced Java and Python engineers who can build and maintain reliable APIs and services.
Guess what? The market is bone dry.
In 2024-25, everyone pivoted to upskilling in AI. Boot camps shifted curriculum. Computer science graduates chased the shiny object. Junior developers who would have been cutting their teeth on REST APIs and database optimisation were instead spending their evenings fine-tuning LLMs.
And I get it. The money was there. Still is, to some extent.
But here's the problem: basic backend engineering skills aren't optional. They're the foundation everything else is built on. Even your fancy AI systems need someone who understands how to make them scale, how to handle data flows, how to build resilient infrastructure.
Beyond Tick Boxes: Diversity Recruitment Strategies That Actually Transform UK Workplaces
Master the Virtual Hot Seat: 7 Video Interview Techniques Recruiters Don't Tell You
How to Master 'Tell Me About Yourself' Interview Question: UK Expert Insights
What recruiters are missing
Last week I spoke to a hiring manager at a health tech startup in Birmingham. They've had a backend engineer role open for 93 days. Ninety-three! They've interviewed 17 candidates. Not one had the right balance of skills.
Finding engineers who can write basic business logic is easy; finding engineers who understand how that logic behaves under load is rare. In 2026, tech leaders need backend specialists who can diagnose connection pool exhaustion, optimize slow PostgreSQL/MySQL execution plans, manage Redis cache invalidation strategies, and design idempotent API endpoints that survive network partitions without duplicating data or corrupting state.
The mid-level squeeze
What's particularly concerning is the thinning of the mid-level band. We have plenty of junior developers (albeit with gaps). We have senior engineers with 10+ years of experience commanding £120K+ salaries.
But that crucial 3-5 year experience bracket? It's looking sparse. Really sparse.
In my recruiting database, applications for mid-level backend roles have dropped 41% since 2024. That's not a statistical anomaly. That's a crisis in the making.
Some of these developers jumped to frontend (where React and Next.js expertise still commands decent salaries). Some moved into DevOps or platform engineering. And yes, some successfully transitioned to AI/ML roles.
But it's left a gap that's becoming impossible to ignore.
The skills that are hardest to find
So what specifically are tech companies struggling to hire for? Based on the last 30+ reqs I've worked on, these are the skills causing the biggest headaches:
1. Performance optimization
Finding engineers who actually understand how to profile and optimize backend services is like finding gold dust. Too many candidates can build something that works, but not something that works efficiently at scale.
2. Distributed systems expertise
Microservices are still the dominant architecture pattern for new builds, but surprisingly few candidates truly understand the challenges of distributed systems. Debugging, observability, eventual consistency - these concepts still trip up most candidates in technical interviews.
3. Legacy system modernisation
Nobody wants to talk about it, but a huge percentage of UK businesses are still running critical systems on older tech stacks. Finding engineers who can bridge between legacy codebases and modern architectures is increasingly difficult.
4. Database performance
This one shocks me. Database skills have declined precipitously. Most candidates under 30 have never written a complex SQL query without an ORM. Understanding indexing strategies, query optimization, and when to use different database technologies is becoming a rare skill.
What can we do about it?
If you're a hiring manager or recruiter, my advice is brutally simple: adjust your expectations or your compensation.
If you're paying £60-70K for a mid-level backend developer in London and wondering why you can't find anyone, stop wondering. The market rate has shifted. Those candidates are getting £85-90K offers elsewhere.
Salary benchmarks from 2023 are ancient history. The developers who would have taken those roles have either upskilled into higher-paying specialities or moved to companies that recognised the changing market.
I've also seen smart companies addressing the gap by:
- Creating deliberate upskilling paths for promising junior developers
- Bringing back structured mentorship programs (remember those?)
- Splitting complex backend roles into more manageable components
- Looking beyond the usual London talent pool to regional cities
But fixing this won't happen overnight. We're looking at a 2-3 year correction period, minimum.
Meanwhile, while everyone keeps obsessing over the AI talent war, I'll be over here trying to find someone who can write a damn database migration that doesn't bring down production.
And trust me - those engineers are becoming just as rare as your prompt engineering unicorns.
Where to focus
If I was giving advice to someone entering the industry today, I'd tell them to double down on backend fundamentals. Yes, learn enough AI to be dangerous. But master the core skills that make systems work reliably at scale.
The pendulum will swing back. It always does.
In my 15 years of tech hiring, I've seen this cycle repeat with mobile developers, frontend specialists, DevOps engineers, and now AI experts. The fundamentals always come back into demand because, well, they're fundamental.
I've got a Java developer role in Leeds that's been open for 4 months. The company finally agreed to increase the salary by £15K last week. They had three qualified candidates apply within days.
Sometimes the skills gap is just a salary gap in disguise.
But I suspect this particular drought is going to get worse before it gets better. Better stock up on backend talent while you still can.
