TentaGen (tentagen.se) is a web-based platform that uses large language models with domain-specific prompt engineering to generate exam questions. The system supports five cognitive taxonomy frameworks (Bloom's simplified and full, NBME, SOLO, Feisel-Schmitz), multiple question types (MCQ, single best answer, true/false, matching), and export formats compatible with established examination platforms including WISEflow and Inspera. Educators specify subject, topic, cognitive level, number of questions, and language. Source material such as course documents, URLs, or recorded lectures can be uploaded to guide the AI toward specific curriculum content. The AI generates and suggests — the educator reviews, edits, and approves every question before use. The platform also supports exam-specific presets that can be configured to align with the quality criteria of specific examination organisations or national standards. Additional export formats, taxonomy frameworks, languages, and examination standards can be integrated on request, as the platform is independently developed and maintained by the applicant.