Organizations defining their roles is certainly a challenge. Quite a few roles don't know what their focus is and other roles constrain their search to the poin

Organizations defining their roles is certainly a challenge. Quite a few roles don't know what their focus is and other roles constrain their search to the point that the job requirements are practically a circular dependency. I'm on the research side of the equation which falls under the data science umbrella which is partially covered by ML engineer, but not fully IMO. Personally, working with data there's a ton of older EE derived DSP+image processing ideas which are great but the recruitment process overlooks with how the roles get built. Having the solid math background is something that seems lacking from anyone exposed only to very high level frameworks or bootcamps.