App Ideas for Students: 13 Projects That Solve Student Problems and Look Good on a CV

By Post your build ideas · 30 October 2026 · 16 min read

The best app ideas for students are small enough to finish in a weekend or a semester, solve a problem you have on campus, and leave you with something to show: a live link, a few real users and a README that explains your decisions. Good picks include a house bills ledger, a timetable merger and a group-project contribution log. Choose by what the project proves, not by how original it sounds.

What makes a student project impressive to a reviewer?

A lecturer marking coursework and a person reading your portfolio want different things, but both are asking the same question: can this person finish something and explain it? A third app idea that nobody has seen before matters less than evidence that the project exists, works and was thought through. There are no hiring statistics behind this list. It is practical advice about what a reader can check in two minutes.

A deployed URL that opens. A link beats a screenshot, and a screenshot beats a description. A reviewer who has to clone a repository and install a database will often stop there.
A few real users, even five. Your flatmates, your society, your course group chat. Five people who used it twice tell you more than a project nobody has tried. Write down what they did and what broke.
A README that records decisions. Not “how to install”. Write why you chose this database, what you left out on purpose, and what you would change with another month.
Tests on the part that can be wrong. Pick the one function where a bug costs something: the money split, the clash detection, the grade maths. Ten tests there are worth more than a hundred on buttons.
One thing that is hard on purpose. Authentication with roles, a third-party API, real-time sync, a messy file format. Each idea below names the hard thing it proves.

Most of the ideas here solve a problem students really have, which helps you in two ways. You are your own first user, and you can explain the project in one sentence in an interview. If you would rather start from a short list of ideas strangers have posted, browse the education ideas on the board or the full idea board.

Which student app ideas fit a weekend or a few evenings?

These suit a first project, a hackathon or a short module assignment. Hours are rough guesses for someone who has built one web app before; add half again if you are learning the framework as you go. Each has a stretch feature, which is what turns a finished assignment into something you would link on a CV.

01 “What do I need in the final?” grade calculator

Enter the weights and marks you have so far, and see the exact mark you need in the remaining assessment to reach each grade boundary.

Student problem: Working out whether a 60 in the exam still gets you a pass, in your head, at midnight
Shows on a CV: Clean business logic with edge cases, covered by unit tests
Scope: Weekend, roughly 6 to 12 hours
Stretch feature: Support several grading schemes (percentages, letter grades, grade points) and a share link that stores the setup in the URL, so no account is needed

Grading rules differ between countries, universities and even modules: some round, some cap, some drop the lowest mark. That mess is the project. Write the rules as data, not as if-statements, and test every boundary. Reviewers who read the tests will see you thought about the edge cases.

02 Exam revision spreader

Enter your exam dates and the topics in each, and get a day-by-day plan that revisits every topic more than once before the exam.

Student problem: Revision plans that fall apart on day three and are never rebuilt
Shows on a CV: A scheduling algorithm with constraints, and a plan that is recomputed when you miss a day
Scope: Weekend, roughly 8 to 15 hours
Stretch feature: Export the plan as a calendar file, and let a missed day push the remaining topics forward instead of failing

Keep the algorithm simple and explainable: spread topics evenly, revisit each at growing gaps, and never place two exams’ heavy days together. A plain rule you can describe in a README is better than a clever one you cannot. Do not claim it improves grades. Say what it does: it makes a plan and rebuilds it.

03 Silent study room with a synced timer

A link you share with friends that opens a shared focus timer, so everyone starts and breaks at the same moment.

Student problem: Studying alone at home with nobody to keep you honest
Shows on a CV: Real-time sync between several browsers, and what happens when one of them drops
Scope: Weekend, roughly 10 to 18 hours
Stretch feature: Show who is in the room without names or accounts, and keep the timer correct when someone joins late or their laptop sleeps

The trap is letting each browser run its own clock. Store the start time on the server and let every client compute the remaining time from it. That one decision is what reviewers look for when they ask how you handled sync.

04 A manual for the flat you’re leaving behind

A page you fill in while you live in a rented flat, with the water shutoff, the cellar key and the neighbour with the spare, handed over as a link when you move out.

Student problem: Every new tenant rediscovers the same ten facts the hard way
Shows on a CV: Content modelling, a read-only share link and a printable export
Scope: Weekend to a week of evenings, roughly 12 to 20 hours
Stretch feature: A read-only link for the next tenant, a PDF export and photos attached to each entry

This one comes straight from the board, where the original post says every flat has ten pieces of knowledge that exist only in the current tenant’s head. The interesting design question is access: the outgoing tenant writes, the incoming one only reads, and nobody needs an account. See the original idea.

Read “A manual for the flat you’re leaving behind” on the board and copy its build prompt →

05 Citation from a DOI or link

Paste a DOI and get a correctly formatted reference in the style your department asks for, ready to copy.

Student problem: Fixing reference lists by hand the night before a deadline
Shows on a CV: Calling a third-party API, handling missing fields and formatting to a standard
Scope: Weekend, roughly 10 to 20 hours
Stretch feature: Add ISBN lookup for books, a second citation style and a bibliography you can paste in whole

Crossref describes its metadata as openly shared through free APIs, so you can build this without a key or a licence deal. For books, Open Library documents an ISBN lookup API. Real records have missing authors and odd capitalisation, so show the user which fields were empty rather than guessing. Check which style your department requires before you pick one.

Which student app ideas are big enough for a semester project or a portfolio piece?

These take weeks, need a database and have a real design question inside them. They are the ones that turn into the “tell me about a project you built” answer. Before you start, run the idea through the MVP scope planner and read how to scope an MVP; the usual way these fail is by gaining a second feature every week.

01 Group-project contribution log

A shared board where each team member logs what they did, with optional GitHub activity alongside, so the team can see who did what before the hand-in.

Student problem: Group work where the split of effort is argued about at the end, from memory
Shows on a CV: Roles and permissions, a GitHub API integration, and a sensible position on measuring people
Scope: Semester, roughly 40 to 70 hours
Stretch feature: Weekly summaries that flag an uneven split early, while there is still time to rebalance

The design trap is counting commits or lines as if they measured effort. They do not, and a reviewer will notice if you pretend they do. Let people log tasks in their own words, show GitHub activity as context and not as a score, and say in the README why. Your lecturer needs to know before you collect data about classmates.

02 Shared house bills ledger

Log who paid for what in a shared flat, and see who owes whom, with the number of transfers reduced to the minimum.

Student problem: Rent, electricity, internet and groceries split by group chat and goodwill
Shows on a CV: Money stored as whole cents, uneven splits and a debt-simplification algorithm with tests
Scope: Semester, roughly 30 to 50 hours
Stretch feature: Recurring bills, a settle-up button and a monthly statement everyone can open

Keep money in integers from day one, because rounding errors in a ledger are the kind of bug a reviewer can find in a minute. Test a three-person case where the split does not divide evenly. Do not add real payments: recording who owes whom is the project, and moving money is a regulated problem you do not need.

03 Timetable merge and free-slot finder

Upload each person’s calendar file and see the times when everyone in a study group is free.

Student problem: Finding a slot when five people have five different timetables
Shows on a CV: Parsing a messy standard file format, time zones and recurring events
Scope: Semester, roughly 30 to 50 hours
Stretch feature: Suggest the three best slots, ranked by how much travel time falls between existing lessons

Calendar exports follow the iCalendar specification, RFC 5545, and the hard parts are recurrence rules and time zones. Use a parsing library for the format and spend your time on the overlap logic and its tests. Process files in the browser so you never hold anyone’s timetable on a server.

04 Turn the tab you have open into three questions tomorrow

Capture what you actually read this week and get three questions about it on Friday morning.

Student problem: Reading a paper or article and remembering nothing by Thursday
Shows on a CV: A language-model API call with a defined output format, a scheduled job and a cost cap
Scope: Semester, roughly 25 to 45 hours
Stretch feature: Ask again at growing gaps, and let the student mark a question as bad so you can see how often the generated ones miss

The original post makes a sharp point: ten questions is homework, three is a habit. That limit is a good scope decision. The part to build carefully is checking the questions against the source text, because a generated question that the reading does not answer is worse than none. Read the idea on the board.

Read “Turn the tab you have open into three questions tomorrow” on the board and copy its build prompt →

05 A calendar for the household that nobody has to maintain

One screen that shows who is home tonight, who is cooking and who is around at the weekend, fed by one tap instead of data entry.

Student problem: A flatshare that settles “who’s in tonight” in a group chat four times a day
Shows on a CV: Privacy-aware design (availability without content), an installable web app and a one-tap input
Scope: Semester, roughly 40 to 60 hours
Stretch feature: Read from calendars people already keep, but only publish “out from 7”, never what or with whom

The author of the original post names the two hard parts: getting information in without effort, and privacy between flatmates who are not family. Both are product questions as much as code, which makes this a good project to write about. The idea on the board has the full reasoning.

Read “A calendar for the household that nobody has to maintain” on the board and copy its build prompt →

06 Campus textbook swap

Students list the books they are done with by ISBN or course code, and the next cohort finds them without paying a shop’s price.

Student problem: Required books that cost a lot for one semester of use
Shows on a CV: Accounts and roles, image upload, search and moderation of user-posted content
Scope: Semester, roughly 50 to 80 hours
Stretch feature: Fill in the title and author from the ISBN with a book-data API, and let a seller mark a book as taken

This is the one marketplace on the list, so it breaks the “works with one user” rule: it is empty until people list books. Limit it to one department or one hall of residence and seed it with your own books. Do not handle payment; let buyer and seller agree a handover. Real users here are a harder test than any feature you build.

07 Society sign-up and QR check-in

A small site where a student society publishes events, members sign up, and the committee scans a QR code at the door.

Student problem: Societies tracking attendance on paper and sign-ups in a spreadsheet
Shows on a CV: Two user roles (member and committee), QR generation and scanning, and a CSV export
Scope: Semester, roughly 30 to 50 hours
Stretch feature: A capacity limit with a waiting list that fills itself when someone cancels

This has the cheapest route to real users on the list: offer it to one society and run its next event through it. A committee that used it for a term is a better reference than any feature. Be careful with attendee names and emails, delete them after the event, and say so on the page.

08 Which of your abandoned repos is closest to finished

Point it at your GitHub account and it ranks the projects you stopped touching by how near they are to shippable.

Student problem: A pile of half-built coursework and side projects, and no idea which one to finish before applying for placements
Shows on a CV: GitHub sign-in, paginated API calls, rate limits and a transparent ranking heuristic
Scope: Semester, roughly 30 to 50 hours
Stretch feature: Ask two or three questions per repo, as the original post suggests, instead of pretending code alone can tell you

The original post says a repository with 4,000 lines and no login is further from done than one with 400 lines and a working payment flow. That is a good ranking rule to test. Run it on your own account first: the best output is a list of what to finish. See the idea.

Read “Which of your abandoned repos is closest to finished” on the board and copy its build prompt →

How do these student app ideas compare on skills, time and difficulty?

Use this table to pick by the skill you want to show. Difficulty is relative to this list and assumes you have built one small web app before.

Thirteen student app ideas by skill shown, time and difficulty
Idea	Skill it shows	Rough time	Difficulty
Grade calculator	Business logic and tests	6–12 hours	Easy
Exam revision spreader	Scheduling algorithm	8–15 hours	Easy to medium
Silent study room	Real-time sync	10–18 hours	Medium
Flat manual	Content model, share links, export	12–20 hours	Easy
Citation from a DOI	Third-party API, standards	10–20 hours	Easy to medium
Group contribution log	Roles, GitHub API, product judgement	40–70 hours	Medium to hard
House bills ledger	Money in cents, debt-simplification algorithm	30–50 hours	Medium
Timetable merge	File parsing, time zones	30–50 hours	Medium to hard
Tab to three questions	Language-model API, scheduled jobs	25–45 hours	Medium
Household calendar	Privacy design, installable web app	40–60 hours	Medium to hard
Textbook swap	Accounts, uploads, moderation	50–80 hours	Hard
Society check-in	Two roles, QR codes, CSV export	30–50 hours	Medium
Abandoned-repo ranker	OAuth, pagination, rate limits	30–50 hours	Medium

Four of these come from the board and are described from the posts as written: the flat manual, the tab-to-questions idea, the household calendar and the abandoned-repo ranker. The others are editorial suggestions, not posts. None of the hour counts are measurements.

How do you turn a class project into a portfolio piece?

A finished assignment is a repository. A portfolio piece is a repository plus evidence. After the mark is in, spend a few more days on the following, in this order.

Deploy it on a free tier and put the link at the top of the README. Check it opens in a private window.
Get five people to use it and write down what happened. Put one honest sentence in the README, such as “five flatmates used it for a month; two stopped after the first bill”.
Add the stretch feature from the entry above. It is the difference between “I did the assignment” and “I took it further”.
Write the decisions section: what you chose, what you rejected, what you would do with another month.
Test the one risky function, and add a short note on how to run the tests.
Remove anything you cannot defend. A half-working feature in the demo is worse than an honest “not built yet”.

README decisions section to copy

## Decisions

Problem: [one sentence, who has it]
Users so far: [number, who, how long]
Stack and why: [choice, and the reason in one line]
Hard part: [what was difficult and how I handled it]
Left out on purpose: [features cut, and why]
Known limits: [what breaks, and when]
Next month: [the first thing I would build]

If you are a student, GitHub Education offers verified students a Student Developer Pack of partner tools. Partner offers change, so read the current list rather than relying on a blog post.

Can you use AI coding tools for a student project?

It depends on whose project it is. For a personal portfolio piece, using an AI coding tool is your call, and the useful rule is that you must be able to explain every file. For assessed coursework, the answer is whatever your course says, and policies differ between universities and between modules. Some forbid it, some allow it with a disclosure, some allow it for certain tasks.

Read the policy before you start, not after. If it is unclear, ask the lecturer in writing and keep the reply.
Disclose what you used if the course allows it: which tool, for which part.
Keep your own commits and notes. A history that shows you working is the easiest way to show the work is yours.
Be ready to explain it. If you cannot walk through your own authentication code or schedule logic without the tool open, you do not own that part yet.
Do not paste course material or other students’ data into a tool unless the rules clearly allow it.

For a portfolio project where the rules allow it, a bounded prompt works far better than “build me an app”. The vibe coding project picker and the vibe coding project ideas guide help you choose a scope an agent can finish, and the same limits help you review what it wrote.

Where do you find more app ideas to build?

Browse the education category for ideas aimed at learners, or the whole board for anything you would use yourself. Every idea has a generated build prompt you can copy into Claude Code, Cursor or Lovable, which is useful as a starting brief if your rules allow such tools. If you want weekend-sized projects outside of student life, see side project ideas.

Found a problem on campus that you will never build yourself? Post it. It is free, needs no account or email, and is permanent once published, so keep names of real people out of it.

Frequently asked questions

What are good app ideas for college students to build?

Good ones solve a problem you see on campus and can be finished in a semester: a shared house bills ledger, a timetable merger, a society check-in, a group-project contribution log. Each gives you a live link, real users to ask about, and one hard technical part to explain in an interview.

What coding projects look good on a resume or portfolio?

Projects with a deployed link, a few real users, a README that records your decisions and tests on the risky logic. Pick one that proves a specific skill, such as authentication with roles, a third-party API or real-time sync, rather than a tutorial clone. One finished, explained project beats five half-built ones.

What app ideas are good for beginners?

Start with logic-heavy tools that need no accounts: a grade calculator, a revision planner, a citation formatter or a flat manual. They fit a weekend, have clear inputs and outputs, and are easy to test. Add a database and sign-in only on the second project, once the first one works.

How long should a student app project take?

A weekend for a first project and roughly four to eight weeks of part-time work for a semester piece. Set the finish line before you start and cut features to meet it. The ideas above include rough hour ranges, which are estimates for someone who has built one web app before.

Do I need real users for a student project?

You do not need many. Five people who use it twice tell you more than a project nobody tried. Offer it to your flat, your society or your course chat, write down what they did and what failed, and put one honest sentence about it in the README.

Is it okay to use AI to build a university project?

Only if your course allows it. Policies vary by university and module, so read yours and ask the lecturer in writing if it is unclear. Disclose any use the rules require, keep your own commit history, and make sure you can explain every part of the code you submit.

Created with AI assistance and edited for this site. Examples are illustrative and do not represent customer research. About the publisher.