Some developers are waiting 3+ months. Others are collecting offers. The data says both can be true.
Hope everyone had a great Easter weekend and got some well deserved extra days off in the UK and anywhere else that had public holidays.
This week’s theme is the thing you’re describing from the ground: the market is starting to feel more candidate-driven for a small top slice, while the majority is still operating in a tight, employer-led market. It looks a lot like income inequality, just applied to job outcomes.
Are we entering a candidate-driven market?
Not broadly, but for standout profiles, yes.
Here’s why the macro still looks employer-favourable:
UK job postings are still 27% below pre-pandemic baseline and wage growth in posted salaries has eased to 4.0% YoY in February.
UK unemployment is 5.2%, vacancies are ~721,000 (Dec 2025 to Feb 2026) and broadly flat.
LinkedIn’s global view: hiring remains 20% below pre-pandemic, and job transitions are at a 10-year low.
That backdrop supports why many good Python devs are taking 3 months or longer.
But now the “top slice” story:
TrueUp data cited this week shows 67,000+ software engineering openings at tech companies, the highest in 3+ years, with open roles up ~30% so far this year.
In the UK, job postings mentioning AI are 127% above pre-pandemic, and ~7.5% of UK job postings mention AI. The highest AI-mention shares are data & analytics (47%) and software development (41%).
So the market is not uniformly candidate-driven. It’s barbell:
Middle: slower loops, more competition, fewer roles, more “no” decisions.
Top end: fewer candidates, urgent demand, and multiple offers.
Why the hiring bar is so high right now
Three forces are stacking:
1) Productivity pressure beats headcount growth
LinkedIn’s 2026 report explicitly frames companies as prioritising productivity over increased headcount, with AI intensifying that pressure and raising the bar for output per worker.
2) AI expectations are spreading into normal roles
This is not just “hire ML engineers.” Hiring Lab shows AI mentions rising across knowledge-work roles, and employers increasingly expecting AI fluency as routine.
3) Applicant volume and risk aversion
When hiring slows and job transitions drop, competition increases. That makes teams more selective and more likely to default to “safe signals” like brand names, publications, competitions, and directly relevant systems experience.
How Python developers can increase the odds of getting more offers
This is the practical bit. The goal is to turn you into someone a hiring team can say “yes” to quickly.
1) Convert your profile into “evidence, not claims”
In your CV and interviews, anchor everything to:
What shipped
What broke
What you measured
What you changed
What improved (latency, cost, error rate, throughput, revenue)
Top candidates do this naturally. Everyone else can learn it.
2) Use referrals properly
LinkedIn reports applicants are 3.6x more likely to get hired if they’re connected to an employee at the company.
Weekly habit: pick 5 target companies, find 1 relevant engineer per company, ask one specific question, build the thread, then request a referral only after a real exchange.
3) Pick a wedge, not a generic title
General “Python Backend Engineer” is crowded. A wedge is something like:
Python + async and event-driven systems
Python + data pipelines and analytics engineering
Python + AI workflow integration (RAG, evals, guardrails, observability)
Hiring Lab’s AI-mention split (41% of software dev postings) is basically telling you where the wedge is heading.
4) Build a two-link proof stack
Two links beats ten buzzwords:
one repo or PR-level artefact
one short write-up (architecture note, postmortem, scaling story)
5) Interview like a senior even if your title isn’t
Most offers are won on “would I trust you with production next week?”
Practice explaining trade-offs clearly, especially:
retries, timeouts, idempotency
caching, backpressure, queues
monitoring and incident response
6) Run fewer processes, but run them better
Barbell markets punish scattershot applications. Pick 10 companies, go deep, tailor properly, and network into them.
7) If you are not getting traction after 3 weeks, change something measurable
Change one of:
target role wedge
proof links
intro message strategy (referrals)
interview reps
Quick Python watch from the last 7 days
FastAPI 0.135.3 released Apr 1, 2026.
uv 0.11.3 released Apr 1, 2026.
Ruff 0.15.9 released Apr 2, 2026.
fastapi-new 0.0.6 released Apr 3, 2026 (one-command FastAPI project scaffold).
So, are we heading into a candidate-driven market?
The macro says “not really.” The top end says “absolutely.” The outcome depends on whether you can show proof fast enough for a high bar, selective market.
Open question to end on: is this barbell effect getting worse in tech, and if so, what fixes it: better candidate signalling, better hiring processes, or something structural like fewer entry points and higher global competition?
Hiring? Contact
Josh Smith
LinkedIn: Josh Smith, Python Recruitment
Email: [email protected]
Phone: 01727 225 552