CAREER PATH · DATA AUTOMATION
We explain new ideas in plain language, practise repeatedly and build useful projects. Basic Python variables, loops and functions are recommended; the first section provides a practical review.
read → validate → transform → report → verifyYour first useful result is only a few careful steps away.EXPLORE BEFORE YOU START
Use Python with CSV, Excel, Pandas, APIs, charts and scheduled workflows to automate repetitive data work.
Basic Python variables, loops and functions are recommended; the first section provides a practical review.
You can read this outline without signing in or completing onboarding. Start with the welcome screen above when you are ready to practise. Use a current browser; a keyboard and larger screen can make longer coding tasks easier, but the lesson interface also adapts to phones.
The in-page coding tasks use pattern checks to confirm required pieces. They do not replace installing the relevant framework, design application, Python environment or hosted service for full execution. Each section includes guided practice, projects and a ten-question exam.
Read an explanation, study the example, try the task and review feedback. Keep your own project files and test them in their intended environment. Completing an activity or a course certificate is not an accredited qualification or a guarantee of employment.
Expand a section to read its chapters and projects. This is the actual course curriculum, not a separate list of promised future topics.
Review Python through file paths, functions and safe repeatable scripts.
Section outcome: Create a command-line automation with clear inputs and outputs.
Locate inputs safely and avoid overwriting the wrong file.
Separate reading, transformation and output responsibilities.
Rebuild and vary the functions and modules example until every important piece feels familiar.
Report enough context to diagnose a failed run.
Keep dates, folders and business rules outside repeated code.
Rebuild and vary the configuration example until every important piece feels familiar.
Copy and classify sample files without deleting originals.
Process an input folder with logs, error handling and a summary.
Read tabular files, preserve important types and write useful workbooks.
Section outcome: Combine monthly files into a checked workbook report.
Handle delimiters, quotes and text encodings without silently corrupting values.
Read and write workbooks while respecting headers, dates and formulas.
Rebuild and vary the excel sheets and cells example until every important piece feels familiar.
Align columns and retain source information across batches.
Produce readable tables without hiding the raw data.
Rebuild and vary the formatting and export example until every important piece feels familiar.
List sheets, row counts and missing headers from a sample workbook.
Validate, combine and export a folder of recurring spreadsheets.
Select, group, join and summarise tables with explicit assumptions.
Section outcome: Produce verified grouped summaries from a real-looking dataset.
Inspect the shape and stored types before calculation.
Select relevant records and aggregate by meaningful categories.
Rebuild and vary the filtering and grouping example until every important piece feels familiar.
Join related datasets while checking unmatched and duplicated keys.
Parse dates deliberately and group by useful periods.
Rebuild and vary the dates and time series example until every important piece feels familiar.
Calculate counts, totals and averages by district.
Explore, join and report a multi-table monthly dataset.
Detect missing, duplicated, malformed and implausible values before reporting.
Section outcome: Create a cleaning pipeline with an auditable rejection report.
Distinguish truly missing data from valid zeros or empty categories.
Define what makes a record duplicate before removing it.
Rebuild and vary the duplicates and identifiers example until every important piece feels familiar.
Reject or flag values outside agreed formats and boundaries.
Record every rule and how many rows it changed.
Rebuild and vary the cleaning logs and reproducibility example until every important piece feels familiar.
Report missing, unique, invalid and duplicate counts by column.
Clean a messy dataset and export clean data, rejected rows and a rule summary.
Retrieve permitted data with authentication, pagination and respectful limits.
Section outcome: Build a cached API importer that survives partial failure.
Check response status and validate expected response fields.
Keep API keys outside source and logs.
Rebuild and vary the authentication and secrets example until every important piece feels familiar.
Follow continuation information and slow down when required.
Record when and where data was retrieved and avoid unnecessary calls.
Rebuild and vary the caching and provenance example until every important piece feels familiar.
Collect mock pages while respecting a rate-limit response.
Retrieve, validate, cache and document an allowed API dataset.
Choose truthful visual encodings and generate readable reports repeatedly.
Section outcome: Create a branded report with charts, tables and plain-language findings.
Match position, length, colour and scale to the comparison being made.
Create readable figures with units, sources and accessible colour choices.
Rebuild and vary the matplotlib and labels example until every important piece feels familiar.
Combine tables, charts and narrative in the format stakeholders use.
State what the data supports and what it cannot prove.
Rebuild and vary the narrative and uncertainty example until every important piece feels familiar.
Create and critique two charts for the same data question.
Generate tables, charts, notes and exports from cleaned data.
Run scripts repeatedly without duplication, silent failure or uncontrolled side effects.
Section outcome: Schedule an idempotent workflow with notifications and recovery.
Make rerunning safe by avoiding duplicated output and irreversible surprise.
Separate the execution schedule from date and folder logic.
Rebuild and vary the scheduling and parameters example until every important piece feels familiar.
Report success or failure to the responsible person with useful context.
Validate transformations and preview changes before writing.
Rebuild and vary the tests and dry runs example until every important piece feels familiar.
Preview renamed or updated files before making changes.
Build a repeatable run with logs, alerts and a recovery guide.
Combine ingestion, cleaning, analysis, reporting and operations for a real stakeholder.
Section outcome: Deliver a maintainable automated monthly reporting pipeline.
Define the decision, source files, schedule and acceptance checks.
Separate ingestion, validation, transformation, reporting and delivery.
Rebuild and vary the pipeline architecture example until every important piece feels familiar.
Compare row counts and totals across stages to detect loss or duplication.
Explain setup, configuration, failures and manual fallback.
Rebuild and vary the documentation and training example until every important piece feels familiar.
Prove that accepted and rejected rows account for every source record.
Automate the full data journey and hand it over with tests and documentation.
Guest progress is saved in this browser on this device. Clearing site data or switching browsers can remove access to that local progress. An optional account can synchronize supported course progress; it is not required to access the lessons. Download or back up project work separately.
Certificates recognize completion of the course activities on this site. They are not university credit, professional accreditation or proof that every program you write is correct.