07-31, 13:50–14:00 (UTC), Red Track
We will talk about the hurdles of trying to convince people at ASML to switch from a very entrenched MATLAB culture to Julia at times when people are just starting to switch to python.
ASML is company in the Netherlands which is the world leader on photo-lithographic system which are crucial for semi-conductor manufacture. For many years its engineers have been using MATLAB to develop hardware and algorithms that are usually rewritten in C/C++ on the main photo-lithography systems. This is the well known two-language problem and for mission critical systems this has been accepted for long time.
For several years ASML has been expanding its market to Analytics and Software for process optimization for which the two-language problem seems to be a little less forgiving than for hardware because time to market is a larger driving force.
For this Julia seems to be a natural candidate to solve the problem. However as this talk will show it is hard to convince "normal" engineers to switch from their trusted tried and truth tools. Moreover, it's even harder to convince them to switch to a relatively unknown/unfamiliar language than a more mainstream like python ... even when it's not the right tool for the job. Which makes things even more difficult is that managers also have a say on this and most think that this is just a fad.
This presentation will give an overview of common reactions, comments and attitudes, and hopefully a guide on how to work around them.
Born in Mexico City. Studied a Bachelors in Chemical Engineering at UNAM. M.Sc. on Materials Science and Engineering at MIT. Studied PhD at TU Eindhoven on Applied Physics.
Worked for Philips Research 1 year.
Working at ASML for 7 years on algorithms.