Home · Personal · Work & Career · Career change & skills
As of 13 August 2026, AI can choose projects for your new-career portfolio.
Most people should hand this to a purpose-built tool.
Can you do it?
5 minutesto a draft.
30 minutesto something you’d act on.
Cost, all in£0
Skill neededchat-fluent
Who has to check ityou
What the alternative costsTeal is an AI job-search workspace for CV tailoring and application tracking, but no price is provided here.
If this goes wrong: you spend weeks building a polished project that does not demonstrate the evidence employers for your target role want.
What to actually do
Use a tool built for this
The route this page recommends
Do it yourself
Second choiceA chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.
How to actually do it
- Open three to five current UK job adverts for the role you want and copy the relevant requirements into a document, removing personal contact details.
- List your existing skills, experience, completed projects, available tools, weekly time, total deadline and any budget, privacy or accessibility constraints.
- Paste the adverts and your list into a chatbot with the copyable prompt, then ask it to rank the five project ideas rather than accepting its first suggestion.
- Compare each claimed skill and evidence requirement in the ranking against the original adverts, deleting any project that depends on tools, data or time you do not have.
- Choose up to two projects and ask the chatbot to turn the preferred one into staged deliverables, a four-week plan and a portfolio case-study outline.
- Build the project, keep dated notes of your decisions and contributions, and test every link, result and factual claim before publishing the portfolio entry.
- Ask someone familiar with the target role to compare the finished entry with the job adverts and identify evidence that is still missing.
Prompt
Help me choose projects for a new-career portfolio. Target role or roles: [insert target role or roles] UK job adverts or person specifications: [paste three to five relevant adverts, with company names and personal contact details removed] Current skills: [list what I can do and my level] Relevant experience: [list work, study, volunteering or personal experience] Projects already completed: [list links or brief descriptions] Time available each week: [insert amount] Total time available before applying: [insert amount] Tools, data and equipment I can access: [list them] Topics I want to work on: [list them] Topics I do not want to work on: [list them] Constraints: [budget, privacy, accessibility, location, collaboration or other limits] Extract the recurring skills and evidence requested in the job adverts. Propose five realistic portfolio projects that can be completed within my stated time and constraints. For each project, give the problem, intended user or audience, deliverable, skills demonstrated, evidence it would create for the target role, estimated build stages, dependencies, risks and a way to show the result publicly without exposing private data. Do not invent employers, datasets, results, qualifications or tools. Mark any assumption clearly. Score each idea from 1 to 5 for relevance to the supplied adverts, feasibility, distinctiveness, evidence quality and learning value. Explain every score. Rank the ideas, recommend the best two, and state what evidence would make me reject each recommendation. Finish with a four-week plan for the first project and a checklist for verifying that the finished portfolio entry contains accurate claims, a working demonstration or suitable substitute, and a clear explanation of my contribution.
Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.
Hand it to a person
The distant thirdA person who owns the outcome does this end to end, worth it when the failure is dear.
What it gets wrong
- AI cannot know which project will make you personally credible to a particular hiring manager beyond the evidence in the material you provide.
- It cannot replace hands-on work, debugging, research, user feedback or the decisions that make a project genuinely yours.
- It can rank ideas using weak or outdated assumptions when the job adverts are vague or your constraints are incomplete.
- It cannot guarantee that a project is distinctive, useful or interesting to the people reviewing your portfolio.
- It cannot verify private context about a workplace, industry or hiring process unless you supply reliable evidence.
Even on a YES, the friction has a name: judgement under ambiguity, context depth and taste.
How we scored this
Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.
| Axis | Score (0–2) |
|---|---|
| Output | 2 |
| Inputs | 2 |
| Verification | 1 |
| Liability | 2 |
| Effort delta | 1 |
| Total | 8 / 10 |
The methodology and its thresholds are published in full.
FAQ
- Can ChatGPT choose the best portfolio projects for me?
- Yes, if you give it the target role, relevant UK job adverts, your current skills and the time and tools you have. It can rank project ideas and explain the evidence each would demonstrate, but you still need to check that the recommendation is feasible and relevant.
- What projects should I put in my portfolio when changing careers?
- Choose projects that create evidence for repeated requirements in the roles you want, rather than projects that only look impressive. AI can suggest options, but the strongest choice depends on your actual skills, available time, interests and ability to explain your contribution.
- Can AI make a portfolio project for me?
- It can help plan, research, write code or text, create test cases and structure the case study. It cannot supply genuine experience, user feedback or honest evidence of work you did not complete, so disclose your contribution and check every claim.
- How do I know if a portfolio project is worth doing?
- Compare it with the requirements in several target job adverts, then check that you can finish a demonstrable version within your time and constraints. A project is a poor choice if it produces little evidence, relies on inaccessible data or cannot be explained clearly in an interview.
Nearby answers
- Can AI create a weekly learning plan for my new career?YES
- Can AI identify gaps in my digital skills?YES
- Can AI plan my route into data analysis?YES
- Can AI prepare me for a career change interview?YES
- Can AI write a personal statement for my course application?YES
- Can AI build a skills-based CV for me?YES
Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.
The newsletter
AI news, new answers and product picks, straight to your inbox.