8 things you need to include in your data science résumé, from one data scientist to another
I frequently have young professionals asking me for advice on how to write a perfect data science résumé. I enjoy sharing my experiences, so I thought I would share a summary of my recommendations.
Note that these recommendations are most relevant to professionals without a lot of data science experience, who are looking to apply for a junior position.
It is impossible (or at least very rare) to know everything about data science, so don’t be surprised if your profile doesn’t fit all the requirements discussed.
These are the 8 things you should include in a strong data science résumé:
- Most relevant experience
- Relevant education
- Actual skills
- Maths / statistics knowledge level
- Coding skills
- Vertical knowledge
- Data management knowledge
- What makes you unique
The overall process: Do your research & set realistic expectations
In an ideal world you would:
- Find an interesting job
- Send your data science résumé
- Pass the interview/assignment with flying colors
- Land the first position you applied for… easy!
In practice, however, you might not even get to the interview. Your CV might get lost in a stack of papers, sandwiched between dozens of similar data scientist résumés.
If you are regularly sending out your CV without ever receiving a call back or interview request, it means your CV is not good enough. It does not mean you are not good, but there might be a better way of presenting yourself, your unique skills and experiences.
When recruiting, reviewers receive hundreds – sometimes thousands of CVs. To streamline the recruitment process, recruiters look for specific items and if they don’t find them right away, your CV will be discarded.
It is important to understand the skills required for a Data Scientist position and challenge yourself to compare them with your experiences. You don’t have to have all of them, but the more you have the more competitive you are. If you don’t know the required skills, scan online job postings to get a better idea.
Before writing your data science résumé and sending it off to the first appealing job posting, be sure to consider the below tips:
1. Read the job posting
It might be obvious, but do not send 10,000 identical CV’s to every data science position you find. It doesn’t work. You have to understand what the company needs. Some Data Science positions are closer to a data architect positions, some are closer to a BI position.
In general, Data Scientist positions can be categorized into three main fields according to the data/techniques:
According to your profile, you might resonate more with one over the other, so focus on your interests.
2. Know the requirements
The daily job of a Data Scientist is quite complex. It might involve gathering data, analyzing data (descriptive analysis), cleaning data/preprocessing, feature engineering, model creation, model validation and productization. You might not be an expert in all of these things.. and that’s alright!
Be smart and try to articulate your experience: where did you achieve success with at least one of the steps?
For instance, you might have some experience writing production code for a project that does not involve any data science and on another project, you did a proof-of-concept to predict something.
Presented together, these two projects demonstrate your ability to do an end-to-end data science project. Some companies are only looking for a subset of those steps, so again it pays to read the job offer.
3. Test your data scientist résumé (iterate)
You should consider the task of writing your data science CV as developing a model. Your first version is your baseline. You measure its performance by checking the ratio number of interviews divided by the number of sent CV.
If one CV scores better, it is your new baseline. Each iteration should provide you with enough information about the performance of your CV to select a new baseline.
The 8 Things You Need to Include in Your Data Scientist Résumé
Now it is time to write your CV. The following points are very subjective and might not be relevant to you. Try to adapt it to your background.
Everything on a data science résumé should be important to the job you are applying to, but it should be ranked from the most to least important.
In my opinion, experience, education, and skills should be first.
1. Experience (Aim for 4-5 bullets/experiences)
It is not necessary to list all the work experience you have.
Try to equate an experience with a project (1 Experience = 1 project) and not an experience with a position (1 Experience = 1 position).
Sometimes you achieved many projects within the same company. In this case, list your projects under the same company name. Each project should contain the challenge and the solution.
If you achieved something in grad school, you can use it as experience:
- Challenge (1-2 lines)
- Design new active molecules for Parkinson’s disease
- Predict how many people will enter a store
- Solution (1-3 lines) – The description of the solution should mention at least one of the following items, otherwise the experience is not qualified.
- Data gathering e.g. collecting data from database ABC
- Preprocessing e.g. Filtering, filling missing values
- Feature engineering e.g. Fourier transform, Wavelet, Autoencoding
- Method used e.g. SVM, KNN
- Validation. Method + metric e.g.using a 10 fold CV, the AUC is 0.78.
- Productization e.g. the solution is currently up and running
This is where you list your degrees, various certifications, etc.
This section is not the list of skills you used in projects but the list of skills you actually have. I am certainly not using 100% of my skills for every project.
The intent of this section is to give the recruiter an idea of your capabilities. Please, do not list everything! Focus on the skills listed in the job offer.
4. Important Statistics
This section is about your knowledge of math/statistics. Keep in mind that the interviewer might ask you to explain them on a whiteboard.
I suggest the following categories:
- Descriptive statistics e.g. : Principal Component Analysis (PCA), Analysis of Variance (ANOVA), Kolmogorov-Smirnov (KS) test
- Unsupervised Learning e.g. K-Nearest Neighbors Algorithm (KNN), Self-Organizing Map (SOM), Hierarchical Clustering…
- Supervised Learning e.g. Linear Regression, Support Vector Machine (SVM), Neural net, Generative Adversarial Network (GAN)
This helps the recruiter assess your level in coding. I am not a fan of stars or numbers to illustrate your level because it is opinionated.
- Languages/libraries e.g. Python, R, Pandas, Keras, scikit-learn, Spark
- Continuous integration e.g. Git, Jenkins
- TDD e.g. Unit tests, integration tests.
6. Vertical knowledge
A list of fields/industries where you applied data science. For instance, smart home, energy, images recognition, etc.
Stay high level!
7. Data management
This section assesses your data management knowledge. You don’t need to know how to set up a spark cluster to use it. In my opinion, Data Scientists are using those tools, they are not maintaining them.
Examples are SQL, Hadoop, AWS s3, etc.
8. What makes you unique
What additional points you want the recruiter to know about you. In other words: what makes you stand out.
Closing Thoughts on the Ideal Data Scientist Résumé
Don’t focus on details. The interviewer will ask you if they are interested. In my opinion, the most important quality of a Data Scientist is common sense, which is already hard to demonstrate on a data scientist résumé… but even harder to demonstrate when I go into too much detail.
It is good to regularly ask yourself high-level questions to check if your current achievements are geared toward the role’s main objectives. You can always highlight this during the interview.
Don’t be afraid of multi-dimensional problems. Your mind can easily understand problems in three dimensions because it can picture the problem. Some people cannot handle a problem with more than three dimensions because they need to represent the problem in their mind to start working on it. As a Data Scientist, you should be able to overcome this limitation.
When you’re writing a data scientist résumé, a one-pager is not always the best idea, but the first page does hold greater importance because it influences the recruiter’s first impression. I would however recommend a maximum of three pages.
Stay up to date
You might also like
Consumer IoT, Industrial IoT
Why IoT needs Machine Learning to (Really) Take Over the World
Putting the Internet of Things and Machine Learning in the same sentence is like playing…
5 IoT Analytics Myths
AI and the IoT have snowballed in attention over the past little while – garnering…
Consumer IoT, Industrial IoT
How 5G will change everything
What is 5G and how is it going to change the world we live in…