Data Science

• The Ultimate Guide to A/B Testing: Lessons from Ronny Kohavi

  May 23, 2024

  I recently watched a video titled The Ultimate Guide to A/B Testing | Ronny Kohavi. Kohavi, a seasoned expert on A/B testing, lays out critical lessons surrounding experimentation culture that resonate well with data-driven organizations. Kohavi emphasizes an essential axiom: “Test everything.” Every code change or feature introduced should undergo an experiment because even small tweaks can yield surprising results. For instance, Kohavi cites examples from Bing where a minor change in ad display led to a stunning 12% revenue increase, equating to around $100 million at the time.

• Building a Data-Centric Culture at DoorDash

  May 15, 2024

  In a recent conversation with Jessica Lachs, VP of Analytics and Data Science at DoorDash, critical insights emerged about building and scaling a data organization that aligns with business objectives. Metrics and Their Impact Jessica emphasizes that analytics should drive business impact rather than merely serve as a support function. Companies often focus on answering “why”—analytics should also address “what do we do now that we know this?” Effective metrics facilitate this process.

• Decoding Financial Times' Next Best Action Recommender System

  January 15, 2024

  In a video, Adam Gajtkowski, a data scientist at the Financial Times, presents the workings of their Next Best Action (NBA) recommender system, which aims to enhance user engagement on their platform. Need for a Recommender System The core rationale behind building a recommender system for a newspaper like the Financial Times is rooted in their business model. With 80% of their customers engaging through digital subscriptions, there’s an abundance of data that can be analyzed to preemptively address user disengagement.

• Building an R Package with LLMs: Insights and Observations

  September 15, 2023

  I just watched a YouTube video by Jared P. Lander, focusing on building an R package with the help of a Language Model (LLM). The video speaks to several intriguing aspects of using LLMs for programming tasks, particularly in R. Basic Function Creation The initial task involved creating a function that counts different file types in a specified folder. When the model was instructed with the prompt, “write a function whose input is a path to a folder and the output is a count of each of the different file types in the folder,” it produced a function using base R.

• From Forecast to Fable: Advanced Time Series Analysis in R

  September 15, 2023

  I watched Mitchell O’Hara-Wild’s talk on Fable, which provided a comprehensive examination of the evolution from the forecast package to Fable in R, highlighting the design decisions that were made to improve time series analysis. Quantitative Focus on Design Decisions Package Evolution: Starting with the TS package at the turn of the century, it brought basic functions such as ARIMA and seasonality filters. The forecast package, introduced in 2003 by Rob Hyndman, advanced this by adding the Exponential Smoothing State Space Model (ETS) and other methodologies.

• Transitioning to Quarto: A Data Scientist's Toolkit

  September 15, 2023

  In a recent YouTube presentation, Daniel Chen discussed his experience transitioning from RMarkdown and Jupyter Notebooks to Quarto. This post distills the core concepts and quantitative elements discussed in the talk. Literate programming allows data scientists to interweave code and prose text within a single document, facilitating better reporting and documentation. Quarto serves as the successor to RMarkdown, bringing multiple enhancements and broader functionality. The Basics of RMarkdown vs. Quarto RMarkdown has been a foundational tool for data scientists, but Quarto builds on its strengths with the following upgrades:

• Wes McKinney - Leveling Up the Data Stack: Thoughts on the Last 15 Years

  September 15, 2023

  Wes McKinney’s talk on YouTube delivers a deep retrospective on the evolution of data science tools over the last 15 years, focusing on key projects including Pandas, Apache Arrow, and their implications for future software development. McKinney kicks off by addressing how Python’s role has expanded in statistical computing, and he notes a contrast to R’s established ecosystem. The primary concern he raises from his early career is fragmentation in the Python ecosystem.

• Making Decisions with Data: A Framework for Effective Analysis

  August 15, 2023

  I recently watched a YouTube video featuring Caitlin Hudon, a data scientist at Figma, discussing decision-making with data. Her framework offers an intriguing look into how to ensure data analysis aligns with the decisions stakeholders need to make. Caitlin introduced an intake form to clarify analysis requests. This form helps establish the parameters of any analysis request by asking critical questions: What decision are you trying to make? What data will guide that decision?

• Data Product Management: Ensuring Value in Data Work

  July 15, 2023

  Recently, I watched a YouTube video featuring Rick Saporta, where he dives into the critical aspects of data product management within data science teams. Saporta emphasizes that the real challenge for data teams often lies not in the data science itself, but in ensuring that their outputs generate actual value. This nuanced perspective prompts an exploration of the hidden pitfalls in predictive modeling and data science initiatives. Understanding the Gap Between Work and Value Saporta shares a personal anecdote about a predictive model developed for the music industry, which boasted an impressive metric: 80% of the top 20 predicted artists actually achieved significant popularity.

• Influencing Product Managers: Insights from Chris Wu

  July 15, 2023

  In Chris Wu’s talk, How to Win Friends and Influence Product Managers, she frames the importance of relationship-building with product managers through a lens of empathy and evidence rather than pure technical data science. This shift in focus is crucial, given that successful product management often hinges on understanding stakeholder motivations and fostering effective communication. The Role of Persuasion in PM Relationships Wu posits that successful interactions with product managers (PMs) revolve around persuasion.

• Leveraging Plumber in Excel for Predictive Modeling in Energy

  July 15, 2023

  Ryan Klein, a data scientist at Continental Resources, delivered a practical talk on integrating R’s Plumber package with Excel to enhance data-driven decision-making in the energy sector. The implications of such a solution extend well beyond the realm of Excel, yet the choice of interface is significant for user adoption. Problem Statement Continental Resources operates approximately 6,000 wells, each requiring an upfront average investment of around $8 million. This staggering figure underscores the need for predictive models capable of determining the viability of these investments based on varying parameters.

• Dissecting NFL Next Gen Stats: Advanced Metrics Driving Football Analytics

  June 15, 2023

  In a recent presentation by Mike Band, senior research and analytics manager at NFL Next Gen Stats, he outlined a robust suite of advanced metrics derived from real-time player tracking data (Watch the full presentation here). This post aims to unpack the key points related to the data models that could reshape our understanding of player performance and game strategy. NFL Next Gen Stats Overview Next Gen Stats employs RFID technology in shoulder pads to capture millions of data points per game—tracking speed, direction, and position at intervals of every tenth of a second.

• The Unseen Impacts of Data Science on Power Dynamics

  May 15, 2023

  In a recent discussion with Columbia University professor Chris Wiggins, a multitude of significant insights emerged regarding the evolution of data science and its implications on power dynamics. The conversation stemmed from a YouTube video that explored how data and data science have rearranged power structures in modern society, particularly in the last year. Rearranging Power Wiggins draws on the notion that data science rearranges power in a way that is both intricate and profound.

• Column Names as Contracts: From dplyr to DBT

  May 10, 2023

  Emily Riederer’s presentation on Column Names as Contracts - Inspired by dplyr, Available in dbt delves into a topic that significantly impacts data usability: how column names can embody contracts that govern data interpretation and manipulation. Riederer emphasizes the common pitfalls of poor column naming conventions: ambiguous abbreviations, inconsistent casing, and the lack of intrinsic meaning that can impede analysis. For anyone who has worked with messy data, this is a familiar frustration.

• Demystifying WebR: Powering R in the Browser

  April 15, 2023

  I recently watched a compelling presentation by Bob Rudis, titled Bob Rudis - Into the WebR-Verse. Rudis breaks down the technical intricacies of bringing R into the web ecosystem using WebAssembly (WASM). This alignment positions R to function natively in browsers, shifting paradigms for data science. Understanding WebAssembly (WASM) WASM serves as a binary instruction format that allows high-performance execution of code in web environments. Introduced in 2015, it aims to bridge the performance gap traditionally associated with JavaScript.

• Understanding Marketing Channel Attribution with Markov Chains

  April 15, 2023

  I recently watched a comprehensive YouTube video by Jessica Duncan, a marketing data scientist, focusing on the intricacies of marketing channel attribution using data science techniques, specifically leveraging R and Markov chains. At its core, marketing attribution aims to determine which customer touchpoints play a pivotal role in converting leads into customers. These touchpoints can span various channels, such as social media, email marketing, or customer service interactions, and can even include untracked influences like word-of-mouth referrals.

• Hamdan Azhar - An Ode to Permissionless Data Science

  March 15, 2023

  In a recent YouTube video, Hamdan Azhar discusses the evolution of data science and the unconventional trajectory that led him into the field. His insights illuminate how the perception and practical use of data science have shifted markedly since it first gained prominence. Azhar recalls a time when his entry into data science was driven more by curiosity than by the tools and buzzwords that now dominate resumes. He critiques the trend where aspiring data scientists focus heavily on listing technical skills like machine learning and Python libraries without articulating the problems they wish to solve.