M05-08 · AI + Data & Decision Science

Professional Practice in Data Science

The capstone integration course for data science professionals. Builds on CORE-08's freelance operations foundation with data-specific professional practice: notebook discipline and reproducibility standards, code review for analysis, building and maintaining stakeholder trust in data, navigating the ethical dimensions of data work, and establishing a data consulting practice. Students complete a multi-week client engagement simulation that integrates every skill from the major.

65 Hours

8 Learning objectives

Create Bloom's ceiling (?)

5 Competencies

What You'll Be Able To Do

Learning Objectives

Objectives

Depth

Maintain notebook discipline across concurrent analyses: meaningful filenames, metadata headers, stripped outputs for version control, execution-order consistency, and clear exploratory-vs-final status indicators Apply
Conduct code review on analytical work — reviewing dbt models, SQL queries, and Python notebooks for data quality assumptions, hardcoded values, missing edge cases, and scalability issues Evaluate
Create reproducible analysis environments: requirements files, data source documentation, assumption logs, and parameter tracking so that any team member can re-run an analysis and get the same results Create
Build stakeholder trust in data through consistent accuracy, proactive error disclosure, transparent methodology, and metric governance — earning the reputation as the person whose numbers are always right Apply
Communicate "no" to bad data requests: insufficient sample sizes, questions the data cannot answer, requests for certainty that doesn't exist, and ML models that would underperform simple heuristics Apply
Evaluate ethical dimensions of data work: privacy implications of analysis (re-identification risks, surveillance potential), bias in algorithmic recommendations, and responsible data stewardship practices Evaluate
Design a data consulting engagement: scope analysis projects with realistic timelines (including data cleaning time), price appropriately ($75-200/hr or fixed-price with data profiling contingency), and deliver reports suitable for both technical and non-technical audiences Create
Analyze career paths in data (IC analyst to staff/principal, management track, specialization in ML/analytics engineering/data engineering) and build a career development plan based on current skills and target role Analyze

Levels: Remember · Understand · Apply · Analyze · Evaluate · Create — highest demands most original thinking.

Core Competencies

What You'll Master

Reproducibility & Discipline

Notebook standards, environment reproducibility, version-controlled analysis, experiment tracking (MLflow/W&B), production-grade documentation.

Analytical Code Review

Reviewing others' SQL, dbt models, and notebooks for quality, correctness, edge cases, and scalability.

Stakeholder Trust

Earning and maintaining the reputation of data accuracy through consistent methodology, proactive error correction, and transparent communication.

Data Ethics

Privacy-aware analysis, bias detection in recommendations, responsible data stewardship, knowing when analysis should not be done.

Data Consulting Practice

Engagement scoping (including data cleaning contingency), pricing models, deliverable formats (notebooks vs. reports vs. presentations), building reusable client-facing toolkits.

Portfolio Artifact

What You'll Build

End-to-End Data Engagement — Student completes a multi-week simulated consulting engagement: receives a client brief (e.g., "churn spiked — why?"), writes an engagement proposal with realistic scope and pricing, profiles and cleans the data, conducts the analysis (statistical methods, segmentation, root cause identification), builds a dashboard, writes an executive report with uncertainty quantification, and presents findings. Includes all project artifacts: proposal, data quality report, analysis notebooks (exploratory and final), dashboard, slide deck, and a retrospective on time allocation (estimated vs. actual, cleaning vs. analysis vs. communication). Also includes a personal career development plan mapping current competencies to a target role.

Tools You'll Use

Industry Tools, Not Toy Projects

Python (full data stack)

pandas, numpy, scipy, scikit-learn, statsmodels, matplotlib, seaborn, plotly for end-to-end data analysis.

SQL (Snowflake/BigQuery/PostgreSQL)

Multi-warehouse SQL proficiency for querying across different client environments and data platforms.

dbt

Data transformation and testing framework for building production-quality analytical models.

Looker / Tableau / Metabase

BI tools for building client-facing dashboards as part of consulting deliverables.

Claude

AI analysis partner for code generation, methodology review, and report drafting with verification.

Bonsai / FreshBooks

Freelance business tools for invoicing, contracts, and engagement management.