I provided the context (problem, industry, requirements), and then I moved into the methodology: data exploration, open communication with the client, application of ML algorithms, reporting, feedback, and revisiting the solution. Eventually, dockerising the final solution and submitting the report.