GITHUB DATA SCIENCE PORTFOLIO
MONDAYS & WEDNESDAYS
6 PM GMT
24 FEB 2027 - 21 APR 2027
DURATION:
9 WEEKS
MONDAYS & WEDNESDAYS
6 PM GMT
Build a public GitHub portfolio of production-grade data science projects, the kind of proof UK hiring managers look for.
Guided by Sagar Vig, Manager, Data Science & Product Analytics at Expedia Group. You will walk away with a polished, recruiter-ready repository that proves your skills.
THIS COURSE IS FOR YOU, IF...
-
You are a junior data analyst or BI analyst
Excel-based reporting limits your ability to demonstrate advanced Python and analytical skills. This course guides you through building modular Python scripts, exploratory notebooks, and machine learning pipelines committed to GitHub. You will publish a structured GitHub portfolio that validates your applied technical skills for data science roles.
-
You are a student or aspiring data scientist
Academic coursework leaves you without recruiter-ready projects to demonstrate applied capability. You will execute an end-to-end data science project using UK open datasets, covering scikit-learn modelling and reproducible environment configuration. You will exit with a live repository featuring structured README documentation and XYZ-format CV bullet points for junior data scientist applications.
-
You are a mid-level data professional
Your commercial data experience is undocumented and invisible to external hiring managers. This GitHub data science for beginners course teaches professional repository architecture, modular code refactoring, and peer review workflows using GitHub Issues and Pull Requests. You will establish an optimised GitHub presence and LinkedIn profile that clearly communicate the technical depth and business impact of your work.
-
You are a software engineer or technical specialist
Your technical background does not translate directly into visible data science outputs. This course provides a structured framework for scoping UK open datasets, training scikit-learn pipelines, and authoring narrative notebooks. You will build a recruiter-ready GitHub repository that repositions your technical foundation for machine learning roles.
Build an end-to-end data science project employers can clone, inspect, and verify.
You will build an end-to-end project using UK open data, transitioning from initial profiling to modular Python scripts, scikit-learn pipelines, and reproducible virtual environments. You will finish with a structured repository featuring embedded charts, Shields.io badges, a tagged v1.0.0 release, and XYZ-format CV bullet points.
Real-time code reviews, dedicated office hours, and 1:1 guidance.
Learn directly from Sagar Vig across 16 live sessions, supported by 30-minute office hours after every class and included 1:1 guidance. You will conduct peer reviews using GitHub Issues and Pull Requests before delivering a live three-minute walkthrough of your completed repository.
You will refactor exploratory Jupyter notebooks into production-ready Python modules with type hints, docstrings, and a clean src/ directory. Each week you practise professional Git workflows, pin dependencies in virtual environments, and run peer reviews using GitHub Issues and Pull Requests. You build disciplined development habits directly inside VS Code and GitHub.
Sagar Vig brings direct hiring criteria from leading data science teams at Expedia Group. You will analyse real UK datasets from ONS and NHS Digital, study repository architecture conventions from tech teams like Airbnb, and evaluate the exact markers of interview-winning repositories. This gives you clear insight into what senior analytics leaders look for during technical screenings.
Your final deliverable is an end-to-end, publicly discoverable GitHub repository featuring modular scripts, embedded EDA visualisations, pinned dependencies, and a tagged v1.0.0 release. You will present your work in a live 3-minute technical walkthrough, translate the project into XYZ-format CV bullet points, and pin it directly to your LinkedIn profile. You leave with verifiable, inspectable evidence of your technical competence.
- Serves as Manager of Data Science and Product Analytics at Expedia Group
- Leads experimentation and strategic insight initiatives for one of the world's largest travel platforms
- Brings nearly a decade of experience across customer analytics and analytics operating models
- Hired and coached data scientists at every stage, from university graduates to career changers
- Combines hands-on technical depth with senior leadership perspective in commercial data science
Review the course roadmap, evaluate technical hiring standards across the UK data landscape, and examine what distinguishes a compelling GitHub portfolio from an abandoned repository of unfinished notebooks.
- Portfolio standards in the UK
- GitHub candidate profile audits
- Class assessment criteria
- Portfolio compounding flywheel
- Case study: Notebook graveyards
- Live lab: Rating live profiles
Master core Git commands and local repository workflows, establishing disciplined version control habits and clear commit conventions that demonstrate professional software engineering standards to prospective technical employers.
- Version control principles
- Core Git CLI commands
- Commit hygiene and conventions
- Portfolio strategy and quality
- Case study: Airbnb monorepo style
- Live lab: Git configuration and push
Assignment #1: Git Practice Repository
Initialise a public practice repository with at least three meaningful commits and one closed Pull Request.
Design a high-impact GitHub profile landing page using custom Markdown, dynamic activity metrics, and visual technology stack badges that establish your professional technical identity for visiting hiring managers.
- Profile repository architecture
- Markdown syntax and formatting
- Shields.io tech stack badges
- GitHub activity stats cards
- Case study: Data scientist profiles
- Live lab: Building a live profile
Assignment #2: GitHub Profile README
Publish a live GitHub profile README featuring a bio, tech stack badges, GitHub stats, and featured projects.
Scope a viable data science project using UK public datasets, establishing a structured directory architecture, problem statement, and proper Git configuration before committing initial code to remote repositories.
- Project ideation criteria
- Problem-Dataset-Insight model
- UK public open data sources
- Repository folder hierarchy
- Gitignore exclusion patterns
- Live lab: Scaffold repo structure
Assignment #3: Project Scaffold Setup
Initialise a structured repository with standard folders, a .gitignore file, and a problem statement README.
Establish robust data ingestion and exploratory workflows, managing sensitive credentials securely with environment variables while documenting column schemas and baseline data quality metrics directly in your project repository.
- Open data ingestion workflows
- Credential security and env files
- Automated data profiling
- Data dictionaries and schemas
- Case study: ONS House Price Index
- Live lab: Notebook data ingestion
Assignment #4: Data Acquisition Notebook
Commit an initial data profiling notebook alongside a README updated with a data dictionary and quality summary.
Construct a narrative exploratory data analysis notebook structured across three analytical stages, producing publication-ready visualisations and plain-English Markdown commentary that explain core dataset findings with complete clarity.
- Narrative exploratory analysis
- Three-act analytical structure
- Visualisation design standards
- Figure exports for documentation
- Case study: UK road safety analysis
- Live lab: Narrative EDA notebook
Assignment #5: Narrative EDA Notebook
Commit a structured EDA notebook with at least two polished charts embedded directly into the project README.
Develop reproducible machine learning pipelines with scikit-learn, evaluating multiple baseline and advanced models while tracking comparative evaluation metrics and analytical limitations systematically within your project documentation.
- Scikit-learn ML pipelines
- Model evaluation metrics
- Baseline versus tuned models
- Markdown results reporting
- Case study: UK energy forecasting
- Live lab: Comparative model pipeline
Assignment #6: Core Modelling Pipeline
Commit a modelling notebook comparing two models with a results table, limitations, and future improvements.
Refactor unstructured Jupyter notebook prototypes into modular, documented Python scripts within a dedicated package directory, applying strict type hints, docstrings, and clean command-line execution interfaces throughout your codebase.
- Modular Python packaging
- Function docstrings and typing
- CLI execution entrypoints
- Script extraction from notebooks
- Case study: Refactoring ETL code
- Live lab: Source module extraction
Assignment #7: Modular Python Refactoring
Create a src directory with two documented, type-hinted modules and refactor notebooks to import functions directly.
Build fully reproducible runtime environments using pinned virtual dependency specifications, documenting reliable setup workflows and testing clean-clone installations to guarantee seamless code execution across different developer operating systems.
- Virtual environments with venv
- Dependency pinning methods
- Setup documentation standards
- Cross-platform conflicts
- Case study: Startup version breaks
- Live lab: Requirements generation
Assignment #8: Reproducible Environment
Commit a pinned requirements.txt file, working virtual environment, and verified setup instructions in the README.
Draft comprehensive repository documentation covering business context, methodology, key findings, and project limitations, structuring clear prose that communicates technical outcomes effectively to both engineering leaders and hiring teams.
- Seven-section README framework
- Plain-English copywriting
- The so-what clarity test
- Collapsible Markdown blocks
- Case study: Product versus academic
- Live lab: Core README drafting
Assignment #9: Structured Project README
Complete all seven core sections of the project README in plain English, explaining problem, data, and methodology.
Enhance repository presentation by embedding exported analytical charts, structured summary tables, and dynamic status badges, creating an engaging visual interface that captures recruiter attention within seconds of landing.
- First-impression repo design
- Shields.io status badges
- Markdown image embedding
- Matplotlib styling rules
- Case study: Visual layout review
- Live lab: Embedding badges and charts
Assignment #10: Visual README Enhancement
Publish a formatted README with Shields.io badges, embedded charts, a results table, and a social preview image.
Organise repository architecture by archiving experimental scratch files, squashing redundant commit histories, and publishing a formal v1.0.0 release accompanied by structured change logs and production-ready distribution notes.
- Interactive Git rebase
- Semantic release tagging
- Root directory hygiene
- Changelog documentation
- Case study: Commit cleanup audit
- Live lab: Tagging release v1.0.0
Assignment #11: Repository Architecture
Finalise repository architecture with archived scratch files, a CHANGELOG.md file, and a tagged v1.0.0 release.
Optimise repository discoverability through curated technical topic tags, concise descriptive summaries, and standardised open-source licensing, ensuring your work indexes prominently across GitHub search and technical recruiter queries.
- Repository discovery metadata
- GitHub topic keyword strategies
- Open-source software licences
- Citation file creation
- Case study: Topic search hiring
- Live lab: Metadata and licence setup
Assignment #12: Metadata and Licensing
Add a repository description, 8 to 10 GitHub topics, an MIT licence file with a badge, and a CITATION.cff file.
Practise professional peer code reviews by auditing classmate repositories against rigorous evaluation rubrics, raising actionable GitHub Issues and providing constructive technical feedback using proven industry collaboration etiquette.
- Industry code review standards
- GitHub Issues bug triage
- Constructive feedback etiquette
- Six-criteria review rubrics
- Case study: Fintech PR review audit
- Live lab: Peer repository audit
Assignment #13: Peer Repository Review
Complete a peer review rubric for a classmate, open three labelled GitHub Issues, and post a review summary.
Implement peer review recommendations systematically through issue-linked commits, debugging path discrepancies and clearing execution outputs across every notebook to guarantee seamless end-to-end reproducibility before final portfolio evaluation.
- Issue triage and prioritisation
- Issue resolution via commits
- Cross-platform path debugging
- Clean kernel execution runs
- Case study: Windows path debugging
- Live lab: Feedback implementation
Assignment #14: Issue Resolution and Video
Close all peer review Issues via commits, re-run clean notebooks, and record a 3-minute Loom walkthrough.
Translate repository achievements into high-impact CV bullet points and optimised LinkedIn features, connecting verified technical artefacts directly to prospective roles to maximise callback conversion from technical recruiters.
- XYZ format CV bullet points
- LinkedIn profile optimisation
- GitHub profile repository pins
- Technical project outreach
- Case study: CV callback turnaround
- Live lab: Profile and pin setup
Assignment #15: Profile and CV Integration
Pin the project repository on GitHub, draft two XYZ-format CV bullets, and write a LinkedIn announcement post.
Deliver a live three-minute technical repository walkthrough to your cohort, demonstrating model methodology, analytical findings, and future improvements while presenting your completed portfolio for formal assessment.
- Technical walkthrough delivery
- Live methodology defence
- Long-term repo maintenance
- UK data science networking
- Case study: Walkthrough evaluation
- Live lab: Final project delivery
Assignment #16: Final Presentation
Deliver a live 3-minute repository walkthrough to the class and submit the final GitHub repository URL.
What our students say
MARKETING DIRECTOR
"It was such a great experience, well worth the course fee which I invested personally - I've learnt so much and feel much more confident in my role.."
"The course at ELVTR was a great investment in my career. The materials are top-notch, and the instructors provided excellent support."
"The knowledge. The teacher is very experienced. He is able to answer our questions in depth, and takes the time to do so."
"The course was very helpful in helping me expand my design toolkits by gaining a wealth of new knowledge and inside expertise to stay at the forefront in an age of rapidly developing technological and software advancements."
Fill in the form to start your ELVTR journey.
We will contact you to clarify all the details.