Topper headline text
AI Career Toolkit
What is AI, actually?
Plain language, no jargon: what these tools are, what they cannot do, and why the ones everybody is talking about arrived all at once.
AI in plain language
“Artificial intelligence” describes software that can recognize patterns, generate writing or images, or make predictions from data, without a person writing an explicit rule for every case. The kind everyone’s talking about right now, tools like ChatGPT, Google Gemini, Microsoft Copilot, and Claude, is called generative AI: it can hold a conversation, draft writing, and answer questions in plain language.
You’ve already been using it
Spam filters, autocomplete, the “recommended for you” row on YouTube or Instagram, even your bank flagging a strange charge, that’s all AI too, just not the conversational kind. Generative AI feels new mainly because you can now talk to it directly, instead of it quietly working in the background.
Why this matters for your career now
Look at job postings in your field right now: many already mention AI tools somewhere in the requirements, even for roles that have nothing to do with tech. Learning the basics now, on your own terms, puts you ahead, whether you’re staying in your field or changing careers entirely.
Career readiness skills for the AI era
Employers increasingly value people who use AI responsibly and well. These are the skills that show you’re ready, beyond the usual resume basics.
Digital literacy
Comfort using AI platforms, knowing their strengths, and knowing when not to rely on them.
Prompting skills
Writing clear instructions that get you accurate, relevant, professional answers.
Critical thinking
Checking AI output for accuracy, tone, and whether it actually fits what you’re going for.
Ethical awareness
Spotting bias in AI tools, respecting data privacy, and staying honest in your career materials.
Adaptability
Keeping up with new tools and trends, and folding the useful ones into how you work.
Responsible use of AI
AI can speed up your career development. Used carelessly, it can just as easily backfire. Keep these principles in mind.
AI and the future of work
Employers are already folding AI into everyday work, from marketing to data analysis to customer service. Learning to use it responsibly now is a career-readiness skill in its own right. AI can draft, summarize, and brainstorm faster than you can alone. It can’t practice a real interview with you, translate your experience into a new field’s language, or tell you honestly when something isn’t landing. That’s what a Merit America career coach is for.
Will AI change your job? Find out in 30 seconds.
Type a job title — the one you have, the one you had, or the one you are aiming for. You’ll get a score from published labor research, and what to do next.
- Nothing to sign up for
How this calculator works
Type your job
We match it to one of 800-plus standardized U.S. occupations.
We pull your score
Every occupation carries a fixed exposure score from published labor research, not a guess and not AI-generated.
You get a next step
We show what your score means and the Merit America track that builds durable, in-demand skills.
Scores combine two peer-reviewed studies: the AI Occupational Exposure index (Felten, Raj and Seamans, 2021) for knowledge and office work, and the Oxford computerisation study (Frey and Osborne, 2013) for hands-on and routine work. We report the higher of the two, so a role counts as exposed if either lens flags it. Exposure measures how much AI could change the tasks in a job. It is not a prediction that the job disappears.
What actually moves your score
Your job title is not what gets scored. Your tasks are. AI takes over some kinds of work fast and stumbles on others, so two people with the same title can land in very different places.
The skills that stay in demand
Exposure is not destiny. The people who get ahead pair what they already do with the skills AI cannot replicate.
Human connection
Building trust, reading a room, and handling the moments that need a real person. AI can sound warm, but it does not feel.
Judgment in the gray areas
Making the call when the data runs out. Weighing trade-offs and deciding what actually matters.
Original thinking
Coming up with the new idea, not the average of old ones, and asking the right question in the first place.
Communication that lands
Turning something complicated into something people can act on, and bringing them with you.
Working with AI
Knowing what these tools do well, where they fall short, and when to trust your own read instead.
Skill categories drawn from Harvard Business School Online’s research on human skills in the age of AI and Harvard economist David Deming’s work on the rising value of social skills.
Knowing which skills hold up is the easy half. The harder half is getting an AI tool to actually help you build them. That is the next two steps, about seven minutes, and it is the difference between a generic answer and a useful one.
Good prompts aren’t a talent.
They’re a habit.
A prompt is just the instruction you type. The gap between a generic answer and a useful one is almost never the tool — it is three things you put in the ask, and you can learn them in one read.
- A real prompt, taken apart line by line
- The one part almost everybody leaves out
- Five popular tips you can safely ignore
What makes a good prompt
Your results depend less on the AI itself and more on how you frame the ask. Here is one real prompt, taken apart.
I’m a warehouse supervisor with 6 years’ experience, applying for a project coordinator role. A hiring manager who has never set foot in a warehouse will be reading it.
Rewrite this bullet point from my resume: “Responsible for daily stock counts.”
Each one should show what I achieved, not what I was responsible for, with a number in it where I have one.
Three versions, one line each.
Success is the one almost everybody leaves out, and it is the one doing the most work. In a controlled trial, 30 people new to prompting were taught to spell out what a good answer would look like; their scores went up 19 percentage points, against under one point for a group taught the usual prompting tips. Saying what finished looks like beats knowing any formula.
Four ways to get a better answer
What almost everybody gets wrong at the start
Neither of these is a sign you are bad at this. They are both just how the thing behaves.
Five popular tips you can safely ignore
There is a lot of advice going around that nobody has ever shown to work. You are not missing a secret.
Longer is not better either. More relevant detail helps. More words do not.
Three things to keep in mind
- Turn off model training first, then give it the real details. On the personal version of all four tools — paying does not change this — your conversation can be used to train the model unless you switch that off, and a sample may be read by a person. Two minutes in settings, or start a temporary chat. Do not water down your prompt to protect yourself; vague inputs are exactly what produces vague answers. Still leave out your Social Security number, and anything you would not want a stranger reading.
- Read everything with a critical eye. AI can be confidently wrong. Use it to draft and brainstorm, not to hand you a final answer you paste and send without checking.
- AI can give you a draft. It can’t sit across from you and tell you if you’re ready. That’s what a Merit America career coach is for. Use these prompts to prepare for that conversation, not replace it.
That is the habit. Next is doing it — five small steps to send your first one and tell whether the answer is any good. If you have not picked a tool yet, have a look at the four first; any of them works, and the free versions are enough.
Now go and write one.
Five small steps to send your first prompt and judge whether the answer is any good, then the check to run before you send anything an AI tool helped you write.
- Works with any tool, free ones included
- A first prompt you can copy as-is
- Six things to check before you hit send
Getting started, step by step
Not sure where to begin? Start small.
Pick one tool
Try ChatGPT, Copilot, Gemini, or Claude. The free version of any of them is enough to start.
Test a simple prompt
Something like: “Rewrite this resume bullet to sound more professional: [paste it].”
Review the output
Keep what works, change what doesn’t, and fact-check anything specific.
Add your own voice
Personalize the wording so it sounds like you. Don’t just copy and paste.
Compare
Put your original next to the AI-enhanced version and decide what’s actually strongest.
Before you submit anything AI helped write
Run through this checklist on any resume, cover letter, or career document AI touched.
Make these prompts yours.
Written for someone changing careers. Search for the moment you’re in, fill your details in once, and every prompt below rewrites itself.
- 51 prompts across 7 categories
- Your details stay in your browser
- Nothing to sign up for
Keep these prompts
We’ll email you the full library, organized by category, so you can keep it and refer back whenever you need it. Unsubscribe any time.
That is the toolkit. If you are weighing whether any of this leads anywhere — what a program costs, whether you have to hand over an email, what the score does and does not mean — those are the questions people actually ask.
Which AI tool should you use?
Every prompt in this toolkit works with any of them, so use whichever you already have. None of these are Merit America products, and none of them pay us to be listed.
Every prompt here works with any of these tools. Use whichever you already have. One rule for all four: on the personal versions, including the paid ones, your conversations can be used to train the model unless you switch that off, and a sample may be read by a person. Paying does not change it. Turning it off takes about two minutes in settings, and every one of these also has a temporary or incognito chat that is never used for training. Do that first, then give it your real details — that is what makes the answers useful.
ChatGPT OpenAI
The all-rounder. If you only try one, start here. It’s strong at drafting, brainstorming, and back-and-forth. The free plan gives you its best model for a limited number of messages a day, then drops to a lighter one.
Reach for it: resumes, cover letters, interview practice, and general career questions.
Google Gemini Google
Best if you already use Gmail and Google Docs. Free with a Google account. It’s good at pulling in current information from the web, so it’s handy for researching a field, a salary, or an employer.
Reach for it: researching careers, pay, and companies when you want up-to-date answers.
Microsoft Copilot Microsoft
Best if you use Word, Excel, or Outlook. Free, built into Windows and the Edge browser, often with no separate account. It can work right inside a document, so it’s useful for polishing a resume you already have open.
Reach for it: help inside the Office programs you already use.
Claude Anthropic
Best for writing that needs to sound human and careful. Free tier available. It tends to write in a natural, less generic voice and handles nuance well, though it has tighter daily limits.
Reach for it: refining the wording of a summary, cover letter, or a tricky email.
Specialized AI tools worth knowing
These go beyond a general chatbot, built for one specific part of the job search. Most have a free tier. None are Merit America products, so use your own judgment on what’s worth paying for.
Gemini Live: talk through an interview out loud and get feedback as you go. Free with a Google account.
Interviewing.io: mock interviews, mostly geared toward technical roles.
CareerFlow: tracks applications and optimizes your LinkedIn profile.
SkillUp’s AI job search tools roundup: a broader list if you want to keep exploring.
Rezi Cover Letter Writer: drafts a first version fast.
Cover Letter CoPilot: another quick-draft option.
Grammarly’s cover letter tool: good for a grammar and tone pass.
The questions people actually ask
Short answers, including the two nobody puts on a page like this: what it costs, and whether you have to hand over an email.
Does a high AI exposure score mean I will lose my job?
Where does this data come from?
Why can’t I find my exact job title?
Do I need to give my email to see my score?
How much do Merit America programs cost?
Keep the prompts you’re collecting
We’ll email you the full prompt library, organized by category, so it’s easy to come back to when you need it.