Not necessarily. Many AI projects you see are thin wrappers that show little skill. What matters is whether your projects solve a real problem well. Add AI only where it genuinely helps, and judge yourself by the depth of your work, not by the labels on other people's posts.
What people tell me
My feed is full of classmates and strangers posting AI chatbots, AI resume tools and AI agents. My own projects are ordinary web or mobile apps with a database and login. They work, and I understand every part of them, but they suddenly feel boring and outdated. I am worried recruiters will skip me because I do not have AI in my portfolio.
A composite of the messages behind this question, with personal details left out.
Key takeaways
- A thin AI wrapper shows less skill than a well-built ordinary application.
- Recruiters and engineers look for depth: real users, clear decisions, clean code, deployment.
- Add AI to a project only when it solves a real problem, and measure whether it works.
- Comparing your private progress to other people's public highlights is always a losing comparison.
- Understanding everything you built is an advantage, not a weakness.
Look closely at what "AI project" usually means
Many of the AI projects you see posted are a text box that sends input to a model API and prints the reply. There is nothing wrong with building one as a learning exercise, but it demonstrates very little: one API call, a prompt and a basic interface. An experienced engineer reviewing that project learns almost nothing about the person who built it.
Your "boring" app with authentication, a database, validation, error handling and deployment shows far more real engineering. You are not behind. You are comparing depth to a label.
What reviewers actually look for
When I look at a student's portfolio, I ask a few questions:
| Question | Why it matters |
|---|---|
| Does it work when I open it? | Shows you finish things and care about the user. |
| Does it solve a real problem for someone? | Shows product thinking. |
| Can the person explain their decisions? | Shows understanding, not copying. |
| Is there anything hard in it? | Data modelling, performance, concurrency, integrations. |
| Is the code readable and tested? | Shows you can work in a team. |
None of these questions mention AI. A project that uses AI can score well on all of them, and so can one that does not.
When to add AI, and how
Add a model when it makes the product genuinely better, not to tick a box. Good reasons look like this:
- Your app has lots of unstructured text, and users need to search or summarise it.
- Users describe things in free text and you need to classify or route them.
- You need to extract structured data from documents or messages.
If you do add it, go one level deeper than most people:
- Handle failures, slow responses and bad outputs gracefully.
- Create a small test set and measure how often the feature gets it right.
- Track cost per request.
- Explain in the README why you used AI there and what you would do differently.
That turns an ordinary AI feature into evidence of judgement.
The comparison trap
Social media shows you the launch post, not the three weeks of struggle, the abandoned version, or the fact that the project was mostly generated and the author could not explain it. You are comparing your full private reality to a curated highlight. That comparison will always make you feel behind, whatever you are building.
A healthier comparison is you against you three months ago. Can you now build things you could not build then? Do you understand more of what you ship? If yes, you are moving forward.
A practical plan for the next month
- Finish and polish your strongest existing project: deploy it, fix the rough edges, write a proper README.
- Identify one real use for AI in that project, if there is one. If there is not, that is fine.
- Build the AI feature carefully, with error handling and a small evaluation.
- Write a short post explaining what you built and what you learned, including what did not work.
You will end the month with a project that is both deeper and more current than the ones making you anxious.
A self-check on your own projects
Open your best project and answer these:
- If a stranger opened the live link right now, would it work without explanation?
- Can you name one design decision you made and the alternative you rejected?
- Is there one genuinely hard part: tricky data, performance, permissions, payments, real-time updates?
- Would you be comfortable walking an engineer through any file they picked at random?
Four yeses beat almost any AI demo in a portfolio. If you answered no to some, those are your next tasks, and they will do more for you than adding a chatbot.
An example of adding AI with judgement
Say you built a small platform where students post questions about their courses. An obvious thin feature is "ask the AI anything". A better one is: when a student types a new question, the system finds similar questions that were already answered and suggests them before posting. That uses embeddings and retrieval, solves a real problem (duplicate questions), and is easy to measure: how often did students click a suggestion instead of posting? You can then write in your README, "Duplicate questions fell by a third after adding suggestions." That single sentence says more about you than ten wrapper projects.
What to say if someone asks why you have no AI projects
In a conversation with a recruiter or engineer, you might say: "My projects focus on building complete, reliable applications. I have used AI tools throughout development, and I am adding an AI feature to my main project where it genuinely helps users. I would rather add it where it makes sense than bolt it on." That answer shows judgement, and experienced engineers tend to respect it.
If you are still stuck
Read How to Use AI as a Companion, Not a Crutch, and if you want project ideas with more depth, browse the FYP ideas library. If you want someone to look honestly at your portfolio, join Sefism and, as a member, book a 1:1 session.
Was this answer helpful?
Read next
- How to Use AI as a Companion, Not a CrutchOne rule decides whether AI tools make you faster or hollow you out: use them for work you could do yourself but would rather not, and for work you are actively trying to understand, never for work you cannot evaluate. Where they genuinely excel, where they quietly damage your skills, and the verification habits that separate engineers who benefit from those who accumulate debt.
- How to Start Your Career as a Software EngineerA practical route into the first software engineering job: pick one language and stop shopping, build things too big for a tutorial, learn the five fundamentals you cannot skip, get your code in front of a human, and apply earlier than you feel ready. Includes a twelve month shape to work to and an honest read on what the current entry level market has changed.
Your situation is not quite this one?
Members get written answers to their own questions, a roadmap built for them and feedback on their projects. Early access is open to X and Instagram followers and university students. Prefer to talk? A free call works too.