People
A decade ago, Data Science was a catch-all that covered plumbing, models, and presentation. Now the roles have become much more specialized.
- Data Engineer (data plumber, model deployer)
- Analytics Engineer (defines usable metrics/features from ingested data)
- AI/ML Engineer (builds models, usually less focus on data cleaning or deployment)
- Data Architect (focus on data and tool structures, usually after lower-level IC experience)
- Data Analyst (focus on counting things and reporting)
- Statistical Researcher (old school company hiring a statistician or data scientist)
- Data Scientist (could be anything, but currently skews towards ML)
- Prompt Engineer (DNE, but soon…)
- Leadership
- Lead Data Scientist (could be anything: first data hire, team lead, architect)
- Data Engineering Manager (people management focus with data eng context)
- Data[.*] Technical Lead (architect and practice focus with data context)
- Head/VP of Data[Science/Eng/etc]
Process
Broad overview of terms
DevOps: The Five Ideals (Gene Kim)
NOTE: Ideal ≠ Culture
- Locality and Simplicity - Making contributions should only require making changes in a single codebase, as opposed to many changes across many codebases.