Students-to-Jobs Platform
A Saudi platform that connects students and graduates with companies and universities — jobs, internships and co-op training, with AI candidate matching.
The problem
In Saudi Arabia, students look for internships and jobs in one place, universities track co-op training in another, and companies sort applicants by hand. This platform puts all three in one place.
For companies, the real problem was volume: a single job could get many applicants, and someone had to read every profile to find the right ones.
What I did
- Built the Jobs and Applications modules end to end, backend and UI.
- Built AI candidate matching: custom scoring services that rank applicants against each job’s requirements.
- Built dashboards for students, employers, universities and admins, with Typesense search across jobs, students and talent pools.
- Delivered the co-op training evaluation system, talent pools, role permissions, OTP login and audit logging.
- Worked in a strict layered architecture (Services, Actions, Repositories, FormRequests) on Docker, with Redis queues for background work.
Architecture
-
Users
- Students & graduates
- Employers
- Universities
- Admins
-
Platform
- Laravel · layered architecture
- Jobs & applications
- AI matching & scoring (key component)
- Co-op evaluation
-
Data
- MySQL
- Typesense search
- Redis queues
The hard part: Rank applicants so the best ones are on page one.
A company posting a job does not want a list in the order people applied. It wants the people who fit the job at the top.
The matching feature compares each applicant’s profile and verified academic background with the job’s requirements, and a custom scoring service turns that into a ranking. The company opens the job and sees its applicants already sorted, instead of reading every profile.
Results
- 4 dashboards: students, employers, universities, admins
- AI applicants ranked against each job
- Live in production