The ATS Challenge
Applicant Tracking Systems are software used by the vast majority of large companies to manage high volumes of job applications. An ATS parses your resume, extracts information, and scores it against the job description. It primarily scans for specific
keywords and phrases related to skills, qualifications, and experience. If your resume doesn't contain the right keywords or is formatted in a way the system can't read, it's often rejected without any human review. This is why a highly qualified candidate might never get a callback; their resume wasn't written for the machine.
What Is Prompt Engineering?
Prompt engineering is the practice of carefully crafting instructions (prompts) to guide an AI model, like ChatGPT, toward a specific and high-quality output. Instead of a simple request like "rewrite my resume," prompt engineering involves giving the AI a clear role, context, and specific constraints to generate tailored content. For resume writing, this means providing your draft, the job description, and a set of precise instructions to refine your bullet points, incorporate keywords, and align your experience with the employer's needs.
A Basic Prompt to Get Started
To begin customising your bullet points, you need a foundational prompt. The goal is to feed the AI your original bullet point and the target job description to get a refined, keyword-rich version. Here is a simple but effective prompt structure you can use: "Act as an expert resume writer. I will provide you with a bullet point from my current resume and the key responsibilities from a job description I am targeting. Your task is to rewrite my bullet point to be more achievement-oriented, incorporate relevant keywords from the responsibilities list, and use strong action verbs. Do not add any information that isn't in my original bullet point. Here is the job description's responsibilities: [Paste 2-3 key responsibilities from the job ad]. Here is my bullet point: [Paste your bullet point]."
Example: Before and After
Let's see this in action. Imagine you're a project manager and the job description emphasizes "end-to-end project lifecycle management" and "stakeholder communication to mitigate risks." Your original bullet point is: "Was responsible for managing projects from start to finish." Using the prompt above, the AI might suggest this revision: "Orchestrated end-to-end project lifecycles, proactively communicating with stakeholders to identify and mitigate potential risks, ensuring on-time delivery." The revised version incorporates the exact keywords, uses a stronger action verb ("Orchestrated"), and frames the responsibility as a proactive achievement.
Advanced Prompting for Deeper Customisation
Once you're comfortable with the basics, you can use more advanced techniques. One popular method is the STAR format (Situation, Task, Action, Result). You can prompt the AI to transform your duties into impact-driven stories. For instance: "Rewrite the following bullet point using the STAR format to highlight a quantifiable result: [Your bullet point]." Another advanced prompt is to ask the AI to analyse the entire job description, identify the top 10 keywords, and then suggest ways to naturally weave them into your experience section. This ensures your resume is dense with relevant terms without sounding robotic or 'stuffed'.
Crucial Final Step: The Human Review
AI is a powerful drafting assistant, not a final author. Relying on its output without review is a significant risk. AI-generated text can sound generic, contain awkward phrasing, or even invent details that aren't true. Always treat the AI's suggestions as a first draft. Your job is to review and edit every line, ensuring it is accurate, reflects your authentic voice, and tells your unique career story. The goal is to create a resume that satisfies the ATS bots but ultimately persuades the human recruiter who reads it next. Never submit a resume without a thorough personal proofread.
