Of late, with most areas of ML being democratized, I am not seeing a whole lot of upside for pursuing a career, specifically leadership, in MLAS over MLE. Traditionally, MLAS leadership roles tend to plateau at VP or Dir of AI etc, but rarely transcends to executive leaderships (CTO or general SVP Eng). Not that MLE leadership (or even generic Data Engineering leadership) transcends easily. But at least comp delta was perceivable. Now even comp gaps are closing fast and approaching an infinitesimal asymptote, with MLAS barely above MLE. definition: MLAS: Includes data scientists and ML scientists working on focussed and applied products (not doing pure research) MLE: Includes MLOps platform engineers, ML engineers, Data engineers working under/supporting ML Ops (excluding data engineers working for simple BI etc) non-niche tech startups: Companies that do not have niche ML as their USP (stable diffusion startups, OpenAI, deep mind back in the day etc) #machinelearningengineer #machinelearning #datascience #mlops #cto #exec #career Blind tax: Before my layoff, led teams at last two roles that did hardcore tabular ML (RL, personalization etc) along with MLE/MLops. TC: 310K + paper money. 9 YoE. sorry, the poll is only for US markets, to keep the data less noisy. Please feel free to create a separate one for other locations or sub-US markets, or post in comments.
agree, im a ML researcher and it is a niche role with not much demand compared to MLE.
markets adjust pretty quickly and are based on supply and demand, not opinion of just the supply, seems like you answered the question yourself on the demand side