Zack, I'm very happy for you, and I'm glad your strategy works.
The people who've worked in my research labs have done them over 160,000 times using hundreds of unrelated substrates ranging from somewhat simple aromatics and alkyl amines to benzodiazepine derivatives and alkaloids with very strange aldehydes. Keep in mind that, when you have research group of 9 running 80 reactions per day (parallel synthesis), the last thing you want is DMF as a solvent. It takes too long to remove via SpeedVac.
We were able to get those 160,000 reactions completed in less than 1 year using sodium triacetoxyborohydride in methanol in the presence of molecular sieves. Yields were 30-90%, workups were immensely simple, and all compounds were purified on mass-triggered prep LC/MS robotic work stations.
You pick the conditions to suit the job, and the pH will not ever be a constant. So long as there are no acid-labile functional groups, you're better off running the pH low (~ 3-4) for a faster reaction. The task was to get three diverse biological leads for ion channel related therapeutic indications in less than one year. We managed to get four (4) diverse biological leads, with an analog from one series entering clinical trials after two more years. That compound is now in phase II clinical trials.