Building a Data Science
Portfolio
Opeyemi Fabiyi
Agenda
My Transition and Learning journey
Why you are now a global Talent
Why do I need a Data Science Portfolio?
What kind of Portfolio should I develop?
How to develop your Portfolio (Demo)
Put yourself out there!
Crafting a good Data Science Resume
Questions
My Transition & Learning Journey
Meet Opeyemi
Current :
• Data Scientist - Cowrywise (FinTech)
• Data Science Mentor – Udacity
• Lead Data Community – Young Data Professionals
• Data Science Mentor – Data Science Nigeria (DSN)
Previous :
• Data Science & Analytics Consultant - Deloitte & Touche (Big 4)
• Machine Learning Engineer - Omdena (AI Startup)
• Supply Chain Planning Analyst – Beiersdorf ( Nivea)
• Data Science Intern – Paylater (Carbon)
Education :
• Electrical/Electronics Engineering (1st Class Hon. ) – University of Ibadan
My Data Analytics Journey - Hacks
Define your Discipline &
Learning path and Consistency
area of Interest Stay committed and dogged
Staying Focus to your learning plan
Business Analyst Join a community &
Be wary of information Continuous learning
Data Analyst
overload
be active
Data Scientist Get a mentor / guide
Data Engineer Stay enthusiastic
Machine Learning Engineer Put your work out there
Don’t be humble about you
do
It’s a Marathon not a Sprint
Why you are now a Global Talent!
Insight Driven Organisation (IDO)
Data Literacy and value discovery from data has bring about two(2) type of organisation
• Traditional & Insight Driven
• An insight- driven organisation (IDO)
Building Block of an IDO
Strategy
People Process Data Technology
Why do I need a Portfolio ?
What does this tell you ?
Even for an entry level role, most companies want to have people with a bit of real life experience.
How do you actually get a real life experience?
Develop a Personal Data Science Project / Portfolio
Evidence of your data
science skills
Why do you need a Data Science Portfolio ?
Your only option to build trust in
any Recruiter without a relevant
degree
What kind of Data Science portfolio
should I develop?
▪ Build a portfolio of data science projects around your interests
If you are passionate about your data science projects, the portfolio will shine through when you discuss them with potential employers.
So choose projects you would love to work with and enjoy showing off.
• https://blog.wetalksound.co/posts/how-we-turned-over-18000-pitchfork-com-music-reviews-data-into-interesting-insights
I love projects where people show that they
are interested in data in a way that goes
beyond homework assignments. Any sort
of class final project where you explore an
interesting dataset and find interesting
results… Put effort into the writeup… I
really like seeing really good write-ups
where people find interesting and novel
things…have some visualisations and
share their work.
~ William Chen
( Data Science Manager, Quora)
▪ Build a portfolio of data science projects that showcase other soft-skills such as communication,
problem-solving & structured thinking
A lot of data science is about communication and presenting data.
• https://towardsdatascience.com/exploratory-data-analysis-feature-engineering-and-modelling-using-
supermarket-sales-data-part-1-228140f89298
• https://fabiyiopeyemi94.medium.com/how-severe-are-car-accidents-in-the-us-c86af93ad050
Projects to Explore
• Data Cleaning Project - Take disparate datasets and make sense of them.
• e.g. You can try scrapping the website to extract a data and analyse
• Data Storytelling Project - A data storytelling project demonstrates your ability to extract insights from data
and persuade others.
• End-to-End data project
• Competitive Data Science :- Kaggle , Zindhi etc.
Data Sources for Projects
https://www.kaggle.com/datasets
https://www.data.gov/
https://archive.ics.uci.edu/ml/index.php
https://zindi.africa/competitions
https://github.com/fivethirtyeight/data
https://www.dataquest.io/blog/free-datasets-for-projects/
How to develop a good Data Science
portfolio
Core Technical
• Expertise required - Intermediate
/ Advanced skills
• Somewhat knowledge of basic
software engineering tool will be
required - Version control (Git/
GitHub)
Two types of Data Science Portfolio Development
Story telling / CRISP-DM modelled
portfolio
• Expertise required - Knowledge
covered in the Academy
• Other skills required - Basic
writing skills (This can be
improved upon with practise)
Cross Industry Standard Process for Data
Mining (CRISP – DM ) Framework
Examples
• https://fabiyiopeyemi94.medium.com/datathon2020-predicting-weather-
disruption-of-public-transport-86373a19e6bc
• https://medium.com/@debbyalenkhe/case-study-a-telecommunication-
performance-analysis-15641be83ef5
Example
• https://linktr.ee/JessAnalytics
• https://risenw.github.io/risingodegua
Demo
Building a static page with Github
Objective
• Introduce Github
• How to make an upload on Github using a simple pet project
• Work on a static project portfolio page
Requirements
• Create a GitHub account (www.github.com)
• Basic HTML tags (Cheats can be downloaded
• https://www.markdownguide.org/cheat-sheet/)
• Make a new GitHub repo and initialise the readme
• Go to the settings
• Go to Github pages and select master branch
• Select theme if you would like
• Go to the link at the top of the GitHub pages section
• Make any changes you would like to using the Markdown of the readme
Crafting a good Data Science Resume
What is a Resume?
• Your Resume is YOU!!!
• The recruiter makes a judgement about your technical skills base on what
your Resume portray
• Without a robust RESUME, you may not get an INTERVIEW invite
DS Resume Features
• Basic Information - Name, Job title, Location, LinkedIn
• Skills (Listed) :- Technical (Excel - Intermediate, Power BI, SQL, Python ,
Python for Data Science (Numpy, Pandas, Matplotlib, Machine Learning
with SckitLearn etc.
• Work Experience
• Projects - For Fresh graduates & those transitioning
• Education
• Volunteering / Competitions participated in / Hackathon
Put yourself out there!
Put yourself out there!
Build your Personal Brand
• What comes to People’s mind about you?
• Having an active online profiles can be a good signal for hiring managers.
Generally, when I’m evaluating a candidate, I’m excited to see
what they’ve shared publicly, even if it’s not polished or finished.
And sharing anything is almost always better than sharing nothing.
- David Robinson
Principal Data Scientist at Heap
Don’t be scared or afraid to make that bold step of putting yourself out there
It’s a Marathon not a Sprint
