Not specifically Survey Monkey API, but API's more generally. Generally looking at the docs,
https://developer.surveymonkey.com/api/v3/, presuming you just want to download data into SPSS (and not tinker with other admin stuff).
The approach you should likely take is to bulk download the cases into memory in python (it is a bit tricky, since you need to page through results). And then import that python object into an SPSS dataset.
Some API's it may make sense to do "one row at a time", if you need to rate limit and the data is already in SPSS in some form. And for that SPSSINC TRANS can work. But the Survey Monkey API has such strong rate limits, you need to do the bulk downloading. (It would be much easier if they just exposed a single bulk download data responses end point, but they do not it appears.)
Not sure what you mean by Azure specifically in this context. (I am guessing you mean the CICD dev-ops?) That might be tricky to interact with SPSS, I think you would *need* an SPSS server and submit jobs to the server.