Operational Data Analysis
Python for Operational Datasets I
Load, clean, and document tabular sources with reproducible notebooks and quality checks.
Tuition is informational only. Enrollment staff confirm availability and policies by email.
Request a syllabus callOverview
This opening studio focuses on resilient ingestion patterns for messy operational exports. You will standardize dtypes, profile missingness, and publish lightweight data dictionaries that downstream teammates can trust. Projects mirror cross-org workflow handoffs rather than one-off spreadsheet edits.
What you build
- Notebook templates with typed column contracts
- Automated profiling reports you can attach to tickets
- Chunked reads for larger CSV and Parquet drops
- Git-friendly folder layout for analysis snapshots
- Pair review checklist for schema drift
- Optional Pandera checks for critical fields
- Packaging utilities into importable modules
Outcomes
- Ship a documented ingestion path for a sample operations feed
- Explain trade-offs between eager and lazy evaluation for your stack
- Hand off a notebook another analyst can run without rework
Mentor on record
Haneul Park
Lead Python instructor focused on analyst-friendly tooling and reviewable notebooks.
FAQ
Do I need prior Python experience?
Comfort with spreadsheets and basic logic helps. We start with syntax drills, but the pace assumes you can study between sessions.
Are datasets from my employer included?
No. You practice on curated sample packs. Your workplace data stays under your own policies; we never request uploads of sensitive exports.
What is not covered?
We do not teach proprietary BI suites or managed cloud billing consoles. The focus stays on open Python libraries you can version in Git.
Participant notes
Week two profiling lab finally gave me language to explain why a vendor feed kept breaking our activity log sync sheet.