Stand Out with Unique Data Science Projects for Your Resume
Written on
Chapter 1: Introduction to Data Science Projects
Recently, I had a discussion with a junior colleague from my graduate program about the challenges of securing job interviews. They requested a review of their resume, believing that the absence of relevant keywords or quantifying past experiences was the issue. However, if you are still working on classic projects like Machine Learning with the Titanic Dataset or Linear Regression on the Boston Housing Prices, it’s time to reassess your approach.
The projects section of your resume is critical, especially for recent graduates or individuals transitioning careers, as prior experiences may no longer be relevant. To break into the data industry as a newcomer, demonstrating your skills through innovative data science projects is essential. These projects not only showcase your capabilities but also instill confidence in potential employers about your ability to perform the job.
Here are five significant projects that played a crucial role in my journey, leading to a year-long internship and two full-time job offers.
Section 1.1: The Internship Project
One of the first data science projects I undertook in 2018 was instrumental in securing my internship at PepsiCo.
Problem Statement: Develop a model to pinpoint optimal locations for new restaurant openings, targeting areas with high population density and minimal competition.
Techniques Used: One-hot encoding, feature engineering and selection, segmentation, k-means clustering, geocoding (to obtain geographic coordinates), and data scraping with Python's BeautifulSoup library.
Tools Utilized: Jupyter Notebook for Python programming.
How did this project help me secure the internship? During my initial interview, I was asked to present my projects. This particular project illustrated a substantial portion of the data science lifecycle. I positioned the project within the context of PepsiCo, suggesting it could assist in identifying locations for their next processing plant, warehouse, or bottling facility. By leveraging unique data points, this project could reveal significant opportunities and cost efficiencies for the company.
Fortuitously, PepsiCo was addressing a similar challenge during my interview, which led to an invitation for a second interview with the hiring manager the following day!
Section 1.2: The Portfolio Project
While in school, I was keen to undertake a comprehensive research project that spanned from data scraping to presenting insights via a dashboard. After researching portfolio projects, I approached my professor with a problem statement.
Over a three-month period, I conducted an analysis of tweets regarding COVID-19 vaccinations, exploring various themes such as sentiment, politics, travel, and vaccine approvals.
Problem Statement: Extract insights from unstructured and diverse COVID-19 vaccine-related tweets, utilizing topic modeling and text analytics to visualize sentiment trends and topic popularity.
Techniques Employed: Data scraping from Twitter, data cleaning, semi-supervised CorEx modeling, sentiment analysis, and unsupervised LDA for feature separation.
Tools Used: Python, Tableau, and MS Excel.
A few months later, after overcoming challenges in scraping data, I authored a blog post on Medium titled "How to Scrape Millions of Tweets Using snscrape" to assist others facing similar issues. I linked my GitHub project in the blog, which subsequently led to meaningful connections, including a conversation with a Senior Data Scientist at a startup about leveraging social media for emotion analytics.
This journey underscored the importance of treating each project as a portfolio piece; the knowledge, code, documentation, and presentation skills I developed became invaluable.
Chapter 2: Additional Projects
The first video titled Make Your Data Science Resume Stand Out 2023 discusses strategies for enhancing your resume through impactful projects.
The second video, The Resume That Got Me A Data Science Job, shares insights on how specific projects can lead to job offers in data science.
Section 2.1: Leadership and Consulting Experience
In my role as a technology consultant for a startup transitioning from concept to a viable application, I collaborated with product managers and app developers to secure initial funding.
Problem Statement: Formulate, test, and implement a strategy to launch the Android and iOS application, establishing a distinct market value proposition.
Techniques Utilized: Time-series forecasting, competitor analysis through Tableau dashboards, market research, and information flow architecture.
This comprehensive project experience taught me the value of accountable leadership and the necessity of patience, determination, and resilience in fostering a new business.
Section 2.2: The Full-Time Job Project
My passion for healthcare analytics prompted me to explore the intersection of data and business in the healthcare sector.
Problem Statement: Develop a predictive model for estimating treatment costs based on clinical factors at the time of admission.
Techniques Used: One-hot encoding, feature engineering, statistical t-tests, correlation analysis, logistic regression, and survival analysis.
Understanding the healthcare business context was critical for this project, including the nuances of HMO versus PPO plans and pricing structures.
During my interview, when asked, "Why should we hire you?" the insights gained from this project strengthened my response. My familiarity with industry jargon and rigorous analysis contributed positively to my candidacy.
Section 2.3: Dashboard Projects
Dashboarding is an essential component of data analysis in any organization. While various tools exist, taking inspiration from community projects can be beneficial.
I maintained three Tableau dashboards that I would rotate while applying for jobs, tailoring them to align with specific job descriptions:
- Credit Score Analysis: Evaluated bank loan data to assess default risk.
- Healthcare Expenditure: Compared the costs of medical procedures across hospitals.
- Marketing Promotion Uplift: Analyzed product interactions to enhance engagement and align placement strategies based on key performance indicators.
Every data-related interview typically inquires about experience with dashboarding tools, making it advantageous to rate yourself highly in this area.
Section 3: Tips for Distinctive Projects
- Scrape your own data to provide unique insights.
- Undertake a portfolio project that encompasses the full data science lifecycle.
- Connect project results to business impact to demonstrate value.
- Conduct a data cleaning project and share the dataset openly with the community.
- Develop a data pipeline to showcase your understanding of data workflows.
In conclusion, these projects have significantly impacted my career trajectory. I appreciate your time reading this, and I welcome your thoughts on your data journeys and aspirations for 2022!
Happy Data Tenting!
Rashi is a data enthusiast based in Chicago, passionate about visualizing data and crafting insightful narratives to convey business insights. She works full-time as a healthcare data analyst and shares her knowledge about data on weekends, fueled by a good cup of coffee.