I like the idea of generating several CVs, the minimalist style very much, and also the the no-email-required policy (but see below). Here are some points I se

I like the idea of generating several CVs, the minimalist style very much, and also the the no-email-required policy (but see below). Here are some points I see for improvement: * There is no way to set the order in which publications appear in the CV, default ordering is a bit weird, so this must be customizable. * Latex error messages could maybe be helpful when something goes wrong.. maybe it's some special character I input? * It's nice that it's not tied to an email, but I am a bit reluctant to give money to a place where I could loose access any time. If something goes wrong on your side and you loose my credentials, how will I be able to re-identify? Also, how does this go with GDPR? * Also in this vain.. are answers to secret questions hashed and salted? if they are, maybe this should be stated as the minimalist retro style suggests they are stored plain text. * There are some standards out there that could make your life easier, and also help users interoperate better. I am not an expert, but there is the european CV xml schema[0], as well as the skills and occupations thesaurus[1]. Also, linkedin and such service GDPR::have to offer some export functionality.. it would be nice to be able to import these. * Several national funding agencies have their own lousy CV creation tools, maybe it could be a businesses model to approach them and re-do it for them. Of course, they are interested in other things (e.g. application tracking, scienciometrics, etc), so just thinking out loud here. * There are tools out there that keep track of your publications, and not all allow for bibtex exports. Maybe make crawlers for these? subcontract with the devil (e.g. access scopus)? or make import for the other formats people use for bibliographies (endnote, etc.) * The approach is nice, so it might be a good idea to keep an eye on the pitfalls of other such ventures, namely: * we are stuck with your style decisions regarding templates, we might not like them, they might run out of fashion, or in an attempt to mitigate this you will create millions of templates which are difficult to search for, hell to maintain, and most likely a growing nest of bugs (already my publications won't render with the "awesome" template). * If you want to allow more customization, complexity in the UI will grow (and bugs), and at the end users will be frustrated and just download the data and write their own latex template (or move to another service). * Malicious actors hosting stuff (allowing for uploads without email-verification / captchas) is, imho, playing with fire. The best of lucks! [0] https://europa.eu/europass/en/eportfolio-interoperability https://europa.eu/europass/en/eportfolio-interoperability [1] https://ec.europa.eu/esco/portal/download https://ec.europa.eu/esco/portal/download