Jamie ColeLast year I was one of 1.6 million people on Universal Credit, refreshing Reed and Totaljobs at 7am...
Last year I was one of 1.6 million people on Universal Credit, refreshing Reed and Totaljobs at 7am like it was a religion. I'd send 15 applications a week and hear back from maybe two. My cover letters were technically fine. My CV wasn't embarrassing. But I was getting ghosted constantly.
I started using AI — specifically ChatGPT and Claude — not as some magic wand but as a thinking partner. And the difference wasn't that AI wrote things for me. It's that it helped me stop writing things that sounded like every other applicant.
Here's what actually moved the needle.
When you've been job hunting for a while, you stop seeing your own CV. You know what you meant to say. You stop noticing what the hiring manager actually reads.
I'd been putting "managed social media accounts" on my CV for two years. An AI prompt showed me I was burying the thing that mattered: I'd grown a following from 400 to 12,000 for a small Leicester-based retailer in eight months. Those are two completely different sentences to a recruiter on Reed.
The fix wasn't AI writing my CV. It was AI asking me the right questions.
1. Extracting real achievements from vague job descriptions
Most people list duties. Recruiters want outcomes. I used this prompt constantly:
"Here is my job description for my last role: [paste]. Ask me 10 questions to help me identify specific, quantifiable achievements I can add to my CV. Focus on results, not responsibilities."
This sounds simple. But when you're deep in job search fog, having something force you to think about actual numbers — customers helped, money saved, time reduced — breaks through the noise.
2. Matching your CV to a specific job posting
UK job boards are full of roles that use slightly different language for the same skills. A warehouse supervisor and a "logistics team lead" might be the same job. ATS systems don't always make that connection.
I used this regularly before applying on Totaljobs:
"Here is a job advert: [paste full advert]. Here is my current CV: [paste CV]. List the keywords and phrases in the job advert that are missing or underrepresented in my CV. Then suggest specific edits using my actual experience."
The "using my actual experience" part is important. Without it you get generic filler. With it you get edits you can actually use.
3. Writing cover letters that aren't corporate sludge
Cover letters in the UK are weird. They're expected, but nobody admits to reading them. Except someone clearly does, because the applications with strong cover letters got responses.
My old approach: summarise my CV in paragraph form. My new approach:
"I'm applying for this role at [company]: [paste job advert]. My relevant background is: [2-3 sentences about yourself]. Write a 3-paragraph cover letter that opens with a specific connection to this company or role (not generic enthusiasm), explains what I bring with one concrete example, and closes with a clear call to action. Tone: direct and professional, not sycophantic."
That "not sycophantic" instruction matters more than you'd think. The default AI voice is relentlessly eager. UK hiring managers find it grating.
4. Preparing for competency interviews
Most UK public sector roles and a lot of large private sector ones use competency-based interviews — the dreaded "tell me about a time when..." format. I was terrible at these until I used AI to help me stress-test my answers.
"I'm preparing for a competency-based interview. Here is a likely question: 'Tell me about a time you had to manage a difficult stakeholder.' Here is my draft STAR answer: [paste your answer]. Identify where my answer is vague, where I'm missing impact, and what a follow-up question might expose. Then suggest how to strengthen the weak points."
This is essentially mock interview practice you can do at midnight before a 9am interview. I used it before every interview in my last round of applications and it's the single thing I credit most with finally getting offers.
LinkedIn in the UK behaves differently from LinkedIn in the US. The culture is less loud. Posting your "journey" or "10 lessons from failure" content tends to land badly here. But connection requests with a specific message — referencing something real from their profile or their company — get replied to.
A simple prompt I used for cold outreach:
"I want to connect with [job title] at [company]. Their LinkedIn shows they work on [specific area]. I'm a [your background]. Write a 3-sentence connection request that references their work specifically and explains briefly why I want to connect. No flattery, no selling myself."
Response rates went from near zero to about 30%. Not huge, but meaningful.
After about six weeks of building these prompts by trial and error, I found a pack that had already done most of the work: 100 AI Prompts for UK Job Search — it's £4.99 and covers everything from CV rewriting to salary negotiation scripts. I used some of the prompts in there as a starting point and adapted them. If you're at the start of this process and don't want to spend weeks figuring out what questions actually produce useful output, it's worth the cost of a coffee.
I want to be straight about this because there's a lot of breathless optimism out there.
AI cannot fake chemistry in an interview. It cannot make a hiring manager warm to you over the phone. It cannot fix a genuinely thin work history or explain a two-year gap better than you can in your own words. It cannot apply for jobs for you without that desperation coming through in everything it produces if you let it write everything without your input.
The job search in the UK right now is brutal — rising redundancies, cautious hiring, more competition per role than there was three years ago. AI is a tool for clarity and preparation, not a shortcut around the fundamental difficulty of getting a stranger to believe in you.
But clarity and preparation? That's where most people fall down. And that's exactly where these prompts help.
Good luck out there.