Nostratic is a hypothetical language family consisting of Indo-European, Uralic, Altaic, Kartvelian, Dravidian, Afro-Asiatic, and, according to some authors, miscellaneous other languages. A similar hypothetical family, which may be a subfamily of Nostratic, is Eurasiatic, consisting of Indo-European, Uralic, Altaic, Eskimo-Aleut, and a few other languages.
The majority of historical linguists is doubtful of Nostratic and Eurasiatic. This is because the evidence is considered insufficient, and several mutually exclusive reconstruction attempts are in circulation. The most commonly cited reconstructions are one by the Russian scholars Vladimir Illich-Svitych and Aaron Dolgopolsky (now in Israel), and one by Allan Bomhard and John Kerns. The two teams use different sound correspondences: at most one of the two correspondence sets can be correct - and it may very well be that both are wrong. Clearly, a method that yields so many false positives must be unreliable.
Nevertheless, this gives opportunities for conlanging, and there are indeed Nostratic conlangs. (Of course, an Indo-European or Semitic conlang would be Nostratic by the definition above.) Danny Wier bases Tech on Nostratic, using roots reconstructed by Bomhard; Jörg Rhiemeier’s Hesperic family is related to Indo-European and thus also Nostratic, though the Nostratic theory has little bearing on this project and the author doesn't assume that Nostratic is valid (nor does he assume the opposite).