Career change advice — AI helps map your transferable skills

"I want to switch careers" usually arrives at the same time as a bad Tuesday: the meeting was useless, your manager forgot the project for the third time, and suddenly the idea of becoming a UX designer / product manager / data scientist / anything-else is irresistible. That feeling is real signal. It's also a terrible time to make the decision.

The good career change is boring. It happens over twelve months, with a spreadsheet, with three coffee chats per week, with a portfolio that exists on the internet. This guide is the boring version, which is the version that works.

Signs it's actually time

"Bad Tuesday" is not a signal. "Bad eighteen months" is. The four signs that a real career change is justified:

• You've stopped learning. When the last new skill you picked up at work was over a year ago, you're either at the wrong level or in the wrong field. Both are fixable; only one is by switching.
• The next promotion doesn't excite you. If you imagine yourself one level up and feel nothing, that level isn't the answer. Two levels up usually isn't either.
• You can articulate what you'd switch to. "Something more creative" is not a target. "Senior product designer at a B2B SaaS" is. The clearer the target, the more real the desire.
• You're not running from a single bad situation. If you quit your manager, the field stays the same. If you quit the field, every manager will be a problem.

Three out of four greens means start planning. One or two greens means fix the current job first.

Skill mapping — what actually carries over

Career change panic comes from focusing on what you don't have. Skill mapping reverses that. Take your last three roles and list, for each one, the underlying capabilities — not the job title.

A senior backend engineer switching to product, for example, has: requirements analysis, stakeholder communication, technical estimation, system design, debugging under pressure, on-call ownership, and the ability to read a database. That stack is 60% of a Senior PM job. The missing 40% is roadmapping, user research, and writing PRDs — three things, not thirty.

The fastest tool for this is to feed your resume and the target JD into the AI career consultant and ask: "What from my current experience maps to this role, and what's actually missing?" Five minutes, real list.

The three project test

Once you know what's missing, build three small projects that demonstrate it. Three is the magic number — one looks like a fluke, two looks like a hobby, three looks like a pattern. The projects should be public (GitHub, Figma community, a Notion case study, a blog), and you should be able to explain each one in two minutes.

Projects beat certifications. A hiring manager spends 90 seconds on your resume — they will not parse "Coursera ML Specialization." They will click a link to a working app.

The financial bridge

This is the unsexy section that determines whether the change actually works. Most career changes involve a 10–30% pay cut for the first role, then a recovery of 12–24 months. Plan for that explicitly:

1. Calculate your monthly survival number. Rent, food, healthcare, debt minimums, kids if applicable. Not your current lifestyle — survival.
2. Multiply by 12. That's your bridge target. Most people aim for 6 and run out.
3. Decide if you change while employed or after. Switching while employed is slower but safer. Switching after is faster but the bridge has to be real.
4. Cut fixed costs before you quit, not after. Renegotiating rent or canceling subscriptions when you're already nervous is the wrong order.

The 12-month number sounds dramatic until you've seen a friend take a job they hated because rent was due in 30 days. That's how career changes get reversed.

The network step (the one people skip)

Nobody gets hired into a new field from a cold application. It happens via someone who can vouch that you're not flaky. That someone doesn't exist yet, so you have to build it.

• One coffee chat per week for three months. Twelve real conversations with people in the new field. LinkedIn DM template: "I'm a [your role] thinking about moving into [target role]. I'd love to ask you three specific questions about how the day-to-day works. Twenty minutes by call?" Half will say yes.
• Ask three concrete questions, not "any advice?" Vague questions get vague answers. "What's the worst part of the job that nobody mentions in interviews?" gets useful answers.
• Follow up with one specific thing you did because of the call. "You suggested looking at X — I built Y, would love your take." This is what turns a contact into a reference.
• Don't ask for jobs. Ask for opinions. The jobs come from the same people six months later.

This is the slowest part of the change and the part that nobody Instagrams. It's also the part that converts.

The five mistakes that cost a year

1. Quitting before you have the bridge. The pressure deforms every decision.
2. Going back to school as the default plan. Two years and €40k buys a credential. Six months and three projects buy a portfolio. The portfolio wins.
3. Not building the network in parallel. The job market is invisible from outside. You need ears inside.
4. Applying for senior roles in the new field. The first role in a new field is usually one level below where you were. Resist the title fight. The level resets in 18 months.
5. Letting the old field's identity stay loud. Update your LinkedIn headline to the new direction before you have the role. People should see "designer-in-training" and respond to that.

Career changes look impossible from the start and obvious from the other side. The middle six months are where everyone wants to quit. Plan the bridge, build the three projects, run the twelve coffee chats. Talk to someone who's actually done a level jump or stress-test the plan with the consultant.

Frequently asked questions