I've switched largely to Jupyter / Python for computational linguistics / psycholinguistics because of the pandas / numpy /numba stack, decent off-the-shelf NLP (spacy and gensim), and the ease of moving data into an R kernel for specific analyses and plots. Also nice that any reasonably sized notebook will render on GitHub (and access can be controlled through the accounts system, until something is ready to be public).

One thing I haven't figure out how to do is to generate fully styled LaTeX manuscripts from notebooks (like papaja for RStudio). Is there a way to do this with pandoc?