Skip navigation

How to start with data science?

Embarking on a journey in data science calls for a fusion of skills: 1) applied statistics, 2) IT prowess (think machine learning, coding, SQL), and 3) domain expertise coupled with a hacker's curiosity! A compelling read by Stanford's own David L Danoho titled "50 Years of Data Science" delves into how data science diverges from traditional stats by its focus on learning rather than inference, and its demand for substantial IT and coding chops.

Aspiring data scientists often grapple with the Python vs. R conundrum. I lean towards Python, particularly considering its edge in recent developments like deep learning, though R is no slouch and is making strides. Essential Python libraries to master include pandas & numpy (data wrangling), matplotlib & seaborn (visualization), and sklearn & statsmodel (for machine learning and time-series analysis). Begin with key machine learning pillars: regression and classification. Hands-on practice? Absolutely! Grab datasets from UC Irvine's Repository or OpenML, and embark on a full-cycle data science project, from data cleaning with pandas to applying classification and regression algorithms via sklearn.

You don't need to reinvent the wheel for each algorithm, though. Reference materials like Hastie, Friedman, and Tibshirani's "The Elements of Statistical Learning" are invaluable. To sharpen your statistical skills, especially in Exploratory Data Analysis (EDA), Allen Downey's "Think Stats" is a stellar starting point.

From there, the world's your oyster! Deep learning with Keras, TensorFlow, or PyTorch, interactive visualization, cloud computing via platforms like AWS, Azure, or Google Cloud, or big data analytics with PySpark — choose your path based on your career aspirations.

Breaking into the Field: Strategies for Newcomers

Feeling overwhelmed? A mentor can be a beacon, especially given the breadth of data science. Industries are shifting focus from degrees to skills, which can be both daunting and liberating. Beware of entities exploiting talented individuals with minimal experience, offering underpaid positions or costly yet subpar courses. Remember, quality education doesn’t always come with a price tag. Exceptional free resources are at your fingertips, like Kaggle and Google's ML Crash Course. The real learning? That happens on the job!

PhDs, Degrees, and the Job Market

Hold a PhD or an MS in a quantitative field with some coding under your belt? Dive into self-training and online courses. Top-tier education is available for free; it's your dedication that counts. Practical learning happens in the trenches at work, not just in classrooms. Major corporates like Google and Apple prioritize your skills over your degrees. Know a fantastic resource? Feel free to share, and I’ll spread the word. Reach me at

Acing the Interview

Success hinges on the role and industry. In tech, your resume is a foot in the door, but your interview performance seals the deal. Preparation is key, as many questions are predictable, especially for tech positions. Oil and Gas, however, is a different beast, heavily influenced by market conditions.

Building a Stellar Data Science Team

For companies eyeing the competitive edge, a proficient data science team is no longer optional. Assembling the right squad is an investment, potentially costing millions annually, but the payoff is immense. The landscape is evolving, with a tilt towards open source and cloud computing. Need guidance on strategizing your approach? I’m here to help. Connect with me at

The Path to a Six-Figure Salary

I offer courses and personal mentorship to equip aspiring data scientists with the necessary skills and confidence to nail interviews and secure those coveted positions. Interested? Fill out this form or contact me directly for more details.