Employee attendance is a crucial factor in supporting the smooth operation of an agency. However, manual attendance recording often leads to problems such as inaccurate data, attendance manipulation, and delays in report recapitulation. Based on these problems, this study aims to design and implement an attendance system based on face recognition technology integrated into a website, in order to improve the efficiency and accuracy of attendance recording for non-civil servant employees and interns within the Tegal Regency Transportation Agency. This system was built using the Laravel framework version 12 as the backend and ReactJS as the frontend, with integration through Inertia.js. For the facial verification process, Biznet's face recognition service is used. This system is also supported by a MySQL database to store user data and attendance records. The system development uses the Waterfall method which includes the stages of needs analysis, system design, implementation, testing, and maintenance. The main features in this application include login, face recognition for attendance, user data management, and attendance reports that can be exported in Excel and PDF formats. The implementation results show that the system can function well according to user needs. The use of face recognition technology can speed up the attendance process and minimize the potential for fraud. Thus, this system is expected to be an effective solution in managing employee attendance and support increasing work efficiency in government agencies.