First, congrats on tuplex, super excited to dive in

By way of background - till recently, I was head of Product at OmniSci where we were solving similar problems in large scale data science and came up with our own way (in conjunction with our partners Quansight) to leverage our existing SQL query compiler pipeline (LLVM-based) in OmniSciDB to allow analysts to author python-based UDFs and UDTFs and register the resulting LLVM IR at runtime with the rest of the query compilation workflow. Slightly different motivations (i.e runtime UDFs/UDTFs that can be used within SQL) than for data science, but many similar problems.

On a more mundane note, I tried using the docker/Mac install (Catalina 10.15.7, Docker Mac) - when I run the cell to initialize the tuplex context, the python kernel always restarts and I cant proceed any further. The notebook logs in the console dont reveal too much either, and the log.txt I see is blank.

Any pointers on what I should be looking for?