Thesisai: Redefining academic excellence with structured AI synthesis.

Hi Product Hunt, I'm Daniel, the maker of ThesisAI. About 700 students wrote theses and dissertations with the first version of ThesisAI. Their biggest worry about AI writing tools was always the same: you can't trust what they cite. Invented papers, wrong page numbers, statistics from nowhere. In a thesis, one of those can sink your defense. So ThesisAI 2.0, live today, is rebuilt around one rule: it never cites a source that doesn't exist, and never writes a number it can't trace. - Sources are real. It searches Semantic Scholar, arXiv and Google Scholar live, and confirms every source against CrossRef and OpenAlex. Where the full text is open, your draft cites the actual page. - Numbers are computed, not generated. T-tests, ANOVA, regression, reliability and more run on your own data. A chapter that states a number it can't trace to your results or a source doesn't pass. - Review that reads like a supervisor. A Critic reviews each chapter from three angles and takes your supervisor's feedback into account. A whole-thesis review then checks that your conclusion actually answers your question. - It fits how you already work. Bring your Zotero, Mendeley or EndNote library, write in your own voice, and export to PDF, Word, LaTeX or defense slides in APA, MLA, IEEE, Chicago or Harvard. It's a writing partner, not a ghostwriter. Your supervisor and your institution's rules come first, and every check is there to help you defend each line. Plans start at $19 a month, or there's a one-time Thesis Pass for six months. Citation checks never count against your plan. I'd love your honest feedback, especially if you've written or supervised a thesis: what would make you trust an AI tool with academic work?