Sukarobot Academy, a robotics and programming course provider for children aged 4-18 with branches in Sukabumi, Bogor, and Tasikmalaya, places new students through a face-to-face trial class, a process branch staff report limits scalability, restricts access outside the service area, and depends on individual trainer judgment across all six offered tracks (Robotics, Coding for Kids, Web Programming, Graphic Design, Microsoft Office, and Digital Marketing). This study develops and validates a Sugeno Order-0 fuzzy inference model that formalizes two trainers' placement judgment into 33 IF-THEN rules using age, a dominant-interest-field score, logic score, and prior experience to generate one program-and-level recommendation per candidate. Validation assessed four distinct forms of agreement rather than one validity claim. Computational agreement was perfect: all 25 black-box scenarios and 40 accuracy test cases matched an independent manual calculation. Across 20 hypothetical profiles, the two trainers agreed with each other on 70% of cases; in every disputed case the system matched at least one trainer. Against 8 active students' confirmed learning paths, historical-placement agreement reached 87.5%, with the one mismatch traced to a trainer's in-person judgment of a child's social maturity that the quiz cannot capture. Usability, measured with the System Usability Scale across 12 parents, averaged 79.58 (Good). These results show the system reliably reproduces the sampled trainers' own judgment and is well received by respondents, but should be read as internal consistency with two trainers' expertise rather than independent, population-level validation of placement quality; broader validation across branches and trainers remains for future work.