https://benfeifke.com/Recent content on Call me BenHugo -- gohugo.ioen-usFri, 17 May 2024 09:47:45 +0200https://benfeifke.com/posts/the-two-documents-every-data-scientist-must-prepare-before-interviewing/Fri, 17 May 2024 09:47:45 +0200https://benfeifke.com/posts/the-two-documents-every-data-scientist-must-prepare-before-interviewing/Spoiler alert: neither of them is a resume. Data science interviews can be tough. As data scientists, we can be expected to serve a sort of swiss-army knife function between business intelligence, software engineering, machine learning, data analysis, and product. On top of that, all companies use data scientists differently, and their respective interviews reflect that by spanning a wide array of questions and processes. We data scientists must be prepared to answer questions on statistics, machine learning, ab testing, business intelligence, system design, leetcode style programming questions, ML Ops, and more.https://benfeifke.com/posts/on-choosing-identity/Wed, 10 Apr 2024 09:47:45 +0200https://benfeifke.com/posts/on-choosing-identity/I was walking home from the park today. The road home from my local park splits into two paths shortly before I reach my apartment. There’s the one which is quite wide and cars even drive on sometimes; it’s convenient for carrying bags or groceries, is suitable for bicycles, and it runs directly adjacent to a series of apartment buildings. Then there’s the other path that runs parallel to the first.https://benfeifke.com/posts/feature-engineering-with-latitude-and-longitude/Tue, 26 Mar 2024 09:47:45 +0200https://benfeifke.com/posts/feature-engineering-with-latitude-and-longitude/Leveraging the power in your geospatial data – with code! Many of today’s most competitive tech markets involve points moving around on a map: ride-hailing services (Uber, Lyft, Grab), micromobility services (Lime, Bird), food delivery services (Delivery Hero, Postsmates, Doordash), and more. Moreover, many services that don’t place customers’ locations at the center of their product use-cases still want to know their customers’ locations so that they can better personalize their experiences based on where they are and what’s going on around them.https://benfeifke.com/posts/syntax-comparison-polars-vs-sql/Tue, 19 Mar 2024 09:47:45 +0200https://benfeifke.com/posts/syntax-comparison-polars-vs-sql/Transitioning from Pandas to Polars the easy way – by taking a pit stop at SQL. The secret’s out! Polars is the hottest thing on the block, and everybody wants a slice 😎 I recently wrote a post, “The 3 Reasons I Permanently Switched From Pandas to Polars”, because, well, this is the most common use-case for Polars – as a drop-in replacement for Pandas, for doing single-node data analysis. However, even though this is the most common use-case, transitioning from Pandas to Polars can be a bit strange, given the heavy differences in syntax between the two.https://benfeifke.com/posts/geospatial-indexing-explained/Wed, 03 Jan 2024 09:47:45 +0200https://benfeifke.com/posts/geospatial-indexing-explained/Geospatial indexing, or Geocoding, is the process of indexing latitude-longitude pairs to small subdivisions of geographical space, and it is a technique that we data scientists often find ourselves using when faced with geospatial data. Though the first popular geospatial indexing technique “Geohash” was invented as recently as 2008, indexing latitude-longitude pairs to manageable subdidivisions of space is hardly a new concept. Governments have been breaking up their land into states, provinces, counties, and postal codes for centuries for all sorts of applications, such as taking censuses and aggregating votes for elections.https://benfeifke.com/posts/master-your-target-language-by-conquering-these-8-social-situations/Tue, 26 Sep 2023 09:47:45 +0200https://benfeifke.com/posts/master-your-target-language-by-conquering-these-8-social-situations/“What is language for? It’s not just the naming of things, is it? It’s the lifeblood of a culture, a people.” JRR Tolkien, in the film Tolkien. We all have different reasons for learning a language: for some, a job or an opportunity might have brought them to a country where they don’t speak the language; for others, it could be an effort to learn more about our own ancestry; and still for others, it could be to build intimacy with your spouse’s family.https://benfeifke.com/posts/refactoring-for-data-scientists-a-beginners-guide/Thu, 27 Jul 2023 09:47:45 +0200https://benfeifke.com/posts/refactoring-for-data-scientists-a-beginners-guide/“The only way to go fast is to go well.” - Martin Fowler A Tale of Two Programmers Joe and Jane are university students taking the same “Introduction to Programming” course, and they’ve just been given the final project: build a Tetris clone! Nervous that they will complete such a big project in time, they both go home and get started right away. Joe starts writing code, and he quickly completes the code for the game initiation, and even implements the “L” piece!https://benfeifke.com/posts/the-3-reasons-why-i-switched-from-pandas-to-polars-20230328/Tue, 28 Mar 2023 00:00:00 +0000https://benfeifke.com/posts/the-3-reasons-why-i-switched-from-pandas-to-polars-20230328/I came for the speed, but I stayed for the syntax. At the time of writing this post, it’s been six years since I landed my first job in data science. And, for those entire six years spent doing data science, Pandas has been the foundation of all my work: exploratory data analyses, impact analyses, data validations, model experimentation, you name it. My career was built on top of Pandas!https://benfeifke.com/posts/ecdf-the-only-plotting-tool-a-data-scientist-needs/Tue, 22 Nov 2022 09:47:45 +0200https://benfeifke.com/posts/ecdf-the-only-plotting-tool-a-data-scientist-needs/After reading this article, you’ll never use a histogram again. The modern data scientist swims in an ocean of plotting techniques. From matplotlib to ggplot2; from plotly to d3.js; from histograms, to box plots, to scatter plots, to whatever crazy chart seaborn comes up with next; it can be hard for data scientists to make sense of it all! But worry no longer. Empirical Cumulative Distribution Functions (ECDFs) will take you wherever you need to go; they are the one Exploratory Data Analysis companion you truly need; the one that cuts through all the noise; the one that will always be there to reliably illuminate your dark statistical world of not-yet understood data.https://benfeifke.com/posts/why-are-the-most-common-verbs-irregular-20220904/Sun, 04 Sep 2022 09:47:45 +0200https://benfeifke.com/posts/why-are-the-most-common-verbs-irregular-20220904/Back when I was learning Portuguese, I often found myself wondering something that most language-learners probably wonder at some point: most verbs seem to follow some pattern, but some verbs have some really weird verb conjugations–what’s up with that? What’s more, why is it that the most common verbs are usually the ones with the most weird conjugations? It’s like someone purposefully designed this language to be hard to learn!https://benfeifke.com/posts/how-did-we-count-before-we-invented-numbers-20220624/Fri, 24 Jun 2022 09:47:45 +0200https://benfeifke.com/posts/how-did-we-count-before-we-invented-numbers-20220624/And Then There Were Numbers Sometimes, I like to imagine the moment that the Set of Natural Numbers was invented. It goes a little bit like this: Stage Left: Curt the Caveman walks in to Mack the Caveman’s cave; Mack is picking his teeth with a bone. MACK: “Hey, Curt!" CURT: “Mack! Great to see you. Beautiful day for tooth-pickin’, huh?" MACK: “You bet, Curt. To what do I owe the pleasure of your visit?https://benfeifke.com/about/Wed, 24 Aug 2016 17:51:42 +0000https://benfeifke.com/about/For the past year, I’ve been helping small businesses discover and leverage the immense value locked in their data. Before that, I worked for 5+ years as a data scientist and machine learning engineer for top companies, including the publicly listed food-delivery company Delivery Hero and as the founding engineer at a Y-Combinator backed startup. I’ve worked across a number of domains, including but not limited to: geospatial data and GIS applications, logistics and transportation, healthcare and medical technology, price optimization, social media data, sentiment analysis, building energy performance and modeling, and more.