Optimising Job Applications with LLM

It's easy to feel powerless or lacking in agency during a job search. Suddenly, any position with a couple of appealing keywords catches your attention.

“Should I be applying for that?”

“Do I have the skills or experience?”

“How do I prioritise which roles to apply for?”

This is often compounded by the length of some job application processes. Recently, I grew frustrated with having to recreate information and repeat tasks for each application, and decided to take a new approach. Enter automation.

What if I told you an LLM could assess your suitability for a role, identify areas where you might need to bridge gaps in experience and then generate a ‘packet’  that includes:

• A customised CV intro

• Customised job experience bullet points

• Customised skills list

• A cover letter

• A sourced estimate of a suitable salary range

Aching this has enabled me to significantly reduce my application time, so I wanted to share how this works in case it’s useful for others.

1. Clearly Define Core Experience and Skills

What I Did: For some time, I’ve used an LLM to help me understand complex job descriptions quickly and then summarise them. In parallel, I saw all the repetition I was experiencing with writing unique CV intros and cover letters each time in the tone of voice appropriate for the role. So, I provided a well-defined overview of my professional experience, key achievements, and top skills to act as a 'database' of my professional profile that the LLM could draw from consistently.

Why It Matters: With a core profile set up, the LLM doesn’t need to reinterpret my experience each time. It can use my actual experience to create CV intros, bullet points, and cover letters that reflect my true skills.

2. Establish Targeted Role Preferences and Career Goals

What I Did: I shared information about the types of roles I’m targeting, such as AI and product management in FinTech, as well as my long-term career goals. This includes my technical focus and preferred sectors.

Why It Matters: This orientation helps the LLM tailor each packet to roles aligned with my current focus and future ambitions, highlighting transferable skills for roles outside my direct experience.

3. Save Industry-Specific Language and Sector Keywords

What I Did: By entering various job descriptions, I helped the LLM learn industry language and trends within my target sectors. This supports the LLM in producing outputs that reflect the voice and requirements of those industries, enhancing relevance.

Why It Matters: Using the right sector-specific language, the LLM generates CV elements and cover letters that resonate with hiring managers, strengthening the impression that I’m an ideal fit for both the role and industry.

4. Structure a “Blueprint” for Job Packets Based on Repeated Elements

What I Did: I developed a standard structure for each job packet, including CV intros, role-specific bullet points, cover letters, and even a section for salary expectations.

Why It Matters: This setup reduces the need to ‘start from scratch’ with each application, saving time while maintaining quality. The LLM can then generate all these components at once from a single prompt, each tailored specifically to a given role.

5. Create a Strategy for Bridging Skills Across Roles

What I Did: For roles that aren’t an exact match, since few jobs ever are, I gave the LLM instructions on how to bridge my skills to meet new requirements. For example, it can frame DAM experience in ways that meet the needs of AI or data-driven roles.

Why It Matters: This bridging strategy allows me to confidently pursue slightly different fields or higher-level roles. The LLM effectively ‘translates’ my experience to meet a variety of roles, positioning me as adaptable even if the role would mean entering a new niche.

Summary for First-Time Users

If you’re new to using an AI-powered approach for job applications, the most important first step is to create a core profile of your experience, skills, and target roles. Add industry-specific language from roles you’re interested in, set up a packet template for consistency, and create a bridging strategy to adapt your skills across related fields. With these steps in place, you’ll be able to harness AI to create tailored, impactful applications efficiently.

But be cautious! The responsibility to check each LLM output remains yours. My experience shows that AI is imperfect. It can misinterpret information, mix up experiences between two similar roles, or lack details if I haven’t provided them. And nothing guarantees a job, of course.

So, am I ‘gaming the system’?

I say no. With modern job applications being increasingly time-consuming, job seekers need innovative ways to avoid repetitive work while creating tailored, effective applications. If there’s a tool that offers that potential, it’s an LLM.

With these foundations, anyone can start building an optimised system similar to mine. And if you’ve already started, or have insights to share, I’d love to hear your approach.