Agile for Data Science

“Data Science is naturally non-linear and fits well into the Agile Framework. It does tend to lack clear upfront understanding; therefore, flexibility is required to pivot based on findings.

There are some variations in expectations and process including:
More ambiguity in planning due to exploratory nature of data
Not every iteration will deliver an expected result
Fully staffed team must include Scrum Master, Data Analyst, Business Analyst, Data Scientist, and ML Engineer to be successful”

Agile for Data Science
Required Roles
Data Value pyramid
MLOps Agile
CRISP-ML(Q)