Verified Commute Carpool App for University Students and Staff
A final year project that builds a mobile carpool app for university students and staff who commute daily: members verified by campus email match recurring trips along the same route, share fuel costs fairly, and never expose home addresses. Route-overlap matching, privacy by design and a real pilot on your own campus.
Intermediate
A small team, or one strong student willing to learn something new.2 semesters · 3 students
7 to ship
3 optional extrasSuggested stack
What you should ship
- Sign-up restricted to verified university email addresses, with driver and rider profiles and optional same-gender matching
- Recurring commute setup: days, departure window, and home area chosen as a coarse zone rather than an exact address
- Matching service that ranks drivers for each rider by extra detour time and time-window fit, using real road routes
- Ride requests, confirmations, weekly schedule view, and a one-tap "running late" or "cancel today" notice
- Fuel cost split calculated from route distance and a published per-km rate, settled outside the app, with a running balance
- Pilot on your campus with at least 20 commuters over 3 weeks, reporting matches made, rides completed and cancellations
- Matching benchmark: latency and quality with synthetic data for 2,000 commuters
If you have time left
- Pickup point suggestions on main roads that cut detour time while keeping homes private
- Multi-campus or office-park mode run by an administrator for a whole organisation
- Estimated fuel and carbon saved per member, shown as a term report
The problem with the daily campus commute
At most universities thousands of students and staff drive or ride in from the same parts of the city at the same times, many of them alone in a car, while others spend a large share of their allowance on ride-hailing or wait for unreliable buses. Informal carpools do exist, organised in WhatsApp groups, but they are hard to find unless you already know the right people, they fall apart when one person's timetable changes, and nobody is comfortable sharing a home address with strangers in a group of several hundred.
Offices and industrial areas have the same pattern. The raw material for carpooling (many people, same destination, same times) is there; the coordination and trust are missing.
What already exists and the gap
- Waze Carpool: matched commuters for cost-sharing rides, not profit; Google shut it down in 2022 as commuting fell during the pandemic.
- BlaBlaCar: cost-sharing for intercity trips booked ahead, not daily short commutes.
- Careem, Uber and inDrive: ride-hailing with paid drivers; convenient but costly for a daily student commute.
- Campus WhatsApp groups: free and trusted by some, but no matching, no schedule and no privacy.
The gap is a closed-community carpool for one campus or organisation, where everyone is verified, matching understands recurring timetables and road routes, and home locations stay private.
What you build
- Driver: sets car seats, recurring departure times and a route from a home zone to campus; accepts riders; posts late or cancel notices.
- Rider: sets days and times; sees ranked drivers with detour and pickup time; requests seats; sees the weekly schedule and balance.
- Admin: approves the university domain, handles reports, sees usage numbers.
The technically interesting part
Matching recurring trips on real roads. For a rider, the cost of a driver is the extra time the driver spends picking them up: route(driver home to rider pickup to campus) minus route(driver home to campus). Computing this for every pair is expensive, so filter candidates first with PostGIS (drivers whose route corridor passes near the rider zone, and whose time window overlaps), then call the OSRM routing engine only for the short list. Ranking must combine detour, time fit and seats left. Show evaluators how filtering cuts routing calls and how match quality compares with a brute-force baseline.
Privacy. Store home locations as coarse zones, share an exact pickup point only after both sides accept, and delete it when the carpool ends. Explain what an attacker with a normal account could learn about another member and how your design limits it.
How to evaluate it
- Matching latency at p50 and p95 for 2,000 synthetic commuters, and routing calls saved by the pre-filter.
- Match quality: average detour of your ranking versus the brute-force optimum on small samples.
- Campus pilot with at least 20 commuters over 3 weeks: matches, completed rides, cancellations and a short survey.
- A privacy review listing what each role can see, tested with a second account.
Suggested plan
- Weeks 1 to 3: survey commuters on campus, agree verification with the university, set up OSRM with your city's map.
- Weeks 4 to 8: accounts, commute setup, zones and data model.
- Weeks 9 to 14: matching service with pre-filter and ranking, benchmarks.
- Semester 2, weeks 1 to 5: requests, schedule, notices, cost split.
- Weeks 6 to 10: campus pilot.
- Weeks 11 to 14: evaluation and report.
Path to a product
The first customer is your own university, which may pay a small yearly fee because fewer cars ease parking and support its sustainability reporting. The same product sells to large employers and office parks per active member. Keep it free for members so adoption is not blocked.
Scope warning
Do not handle payments in the app. Do not build live GPS tracking during rides or turn-by-turn navigation. Do not open it to the public; the closed, verified community is the point.
Learn what you need on Sefism
- Mobile roadmap : Flutter, location and notifications.
- Backend roadmap : APIs, geospatial queries and background jobs.
- System design cheat sheet : filtering, caching and latency ideas for the matcher.
- PostgreSQL documentation : indexes and spatial queries behind the pre-filter.
- k6 documentation : load testing the matching service.
Get help with this idea
The matching design and the privacy model are the two decisions to review early. A mentor can help you cut the scope to what your team can finish, choose between the stack options above, and design an evaluation your examiners will respect. You still build it; the help is on decisions, not code.
New here? Create a free account first.
Ideas and guidance, not finished projects
These are project ideas and scoping guidance, published free for students to use as a starting point. I do not build, write, or sell final-year projects, and I do not complete coursework for anyone. Take an idea, make it yours, and build it.
Not sure this one fits you?
I help students pick an idea that matches their skills and their deadline, then scope it down until it is finishable. Guidance only, never done for you.
Stuck on something specific?
Writing only gets you so far. If you want an answer to your situation rather than the general case, book a session and we will work through it together. Sessions are free for approved Sefism members, and a few slots open each week.
Follow along
New writing, resources and project ideas land here first.