The main activity carried out daily for an individual to earn money is by working. Finding a job that matches our educational background is not easy. Many individuals do not know their abilities and the limited information on job vacancies is also an obstacle for applicants who want to find work. Therefore, a 'toolkit is needed that can provide recommendations on what fields of work are by the educational background in question. The hybrid approach method is to combine collaborative-filtering techniques (decision tree algorithm) and content-based (nearest neighbor algorithm). The decision tree algorithm is used to classify the fields of work while for job recommendations, the nearest neighbor algorithm is used. In the nearest neighbor, the similarity formula is used to calculate the proximity between applicants and job vacancies based on matching the existing weights and attributes. The output generated from this system is a list of job recommendations that match the applicant's educational background.