Avoiding inflated A/B test results
SaaS data science
Gaps in your analysis process
Daylight savings gripes
2020 Python developers survey
Explore by Category
Analytics Dispatch Archive
A weekly email about data, data science, and analytics. Curated by the team at Mode.
Avoiding inflated A/B test results
DataOps unleashed. Data in phulkaris. Date-time handling for R.
SaaS data science
Uber’s data culture. Reducing toxicity in language models. Switching-replication design.
Gaps in your analysis process
Bayes applied articles. Data-proficient organizations. Pandemic data failings.
Daylight savings gripes
Time series forecasting with Torch. Minimum wage changes. Getting data out of Excel files.
2020 Python developers survey
Github + RStudio. Data science is different. CA exodus... not so much.
Picking your next binge watch
NFL simulations. Machine learning reliability engineering. How much stats do you need?
Pie chart en français
Data literacy. AI impact on queer communities. WTF Python?
AI incident database
Data feminism. Flowers in data viz. Sports analytics 101.
Running your data science team
Facial recognition datasets. DS in urban planning. Data wrangling in R.
The case for humility in tech
Intro to Shiny. Cats in your R plots. How to scope down PRs.
Data science portfolios 101
Inauguration palette. Senate generations. Good storytellers.
Why you should do NLP beyond English
Data is plural archive. Anomaly detection with SQL. Google Books API.
Baking with machine learning
Data justice. Level up your reports. Best data-driven stories of 2020.
STEM Ph.D. not required
Real-time ML. Analysis paper trail. Underlying values of ML research.
How bad is your Spotify?
2020 datasets. COVID-19 vaccine inequalities. Core concepts.
Data and its (dis)contents
R tutorials. Kim Eng Ky profile. Data scientist productivity.
Building diverse data teams
Timnit Gebru. Datasheets for datasets. Data science tools at Spotify.
Keyboard shortcuts. The four jobs of a DS. Feature store.
Election night with Biden’s data guru
A custom corpus. Politics in the workplace. What polls and business teams have in common.
We’re all dashboard junkies now
Egos and emotions. COVID-19 event risk assessment tool. Spatial causal inference.
Imposter syndrome in data science
Job hunting. Algorithm decision-making. What data science blogs don’t tell you.
Why you should care about form extraction AI
A stats history lesson. CSS in R. Weird AI Yankovic.
When good data analyses fail
Why have a data science portfolio? Modern data infrastructure. Teaching Python to beginners.
Augment, don’t automate
Bayes rules. Ballot counting. Bias in compressed models.
Before R, there was S
Git for R users. Debunking the Target pregnant teenager myth. Tracking Plandemic.
A tale of query optimization
Decolonizing computational sciences. Machine learning tech debt. Ben & Jerry’s.
Working at Amazon
New: calculated fields! Evil data scientist thought experiment. Data request intake form.
The hardware lottery
Debunking narrative fallacies. Data organization in spreadsheets.
What can data scientists learn from DevOps?
Pycon Africa. Modular data stack. ML testing.
Data team ROI
Improving ETAs. AI ethics. sklearn + Flask + Docker.
Machine learning for agriculture
Big book of R. Biased dataset. Ethical AI at Salesforce.
Preventing a “Meow” attack
New: Python Edge. End-to-end data scientists. Visualization mirages.
Are dashboards dead?
New: Mode Explorations. The whiteness of AI. Python typosquatting.
Algorithmic colonization of Africa
Uber’s data quality monitor. How ML & DS work together. Calculated fields.
Scrum for data science
Categorizing products at Shopify. The most popular high school plays. AI harms.
The future of data governance
Writing alt text for data viz. Applied machine learning. 5 popular window functions.
How to lead through a pandemic
Data Analyst 3.0. Stress during technical interviews. Atlas of surveillance.
The evolution of a ggplot
Data infrastructure at Netflix. Shipping ML projects. The case-death gap.
Linear regression in SQL
NLP with Spacy. Making machine learning actually useful. Adventures in R.
Motherhood in tech
Pull request study. Ethics in NLP. Tips for a PhD.
Decision-making in a time of crisis
The gaps between white and Black America. GitHub actions for data science. Common table expressions.
The facial recognition fallout
Data as protest. Causal inference. Machine learning in production.
Black Lives Matter
Amplify Black voices. Protect Black lives.
Data science dominates fantasy football
Bye, Internet Explorer. Exploring missing values. Pixelate to communicate.
How to evaluate customer health in a pandemic
Ubuntu ethics in AI. Rules-based models. Scalable user privacy.
Data scientists in academia
Fuzzy name matching. Solar’s cheap future. Deleting data across microservices.
Bye, data lake. Hello, data mesh.
Animal Crossing sentiment analysis. Mode improvements. Tricky predictions.
Making sense of COVID-19 models
Boosting A/B test power. Shortcuts of Mode power users. Uncertainty visualization.
Cheating at Scrabble with Python
What developers need to know about databases. NYC sidewalk widths. Unprecedented line charts.
The unemployment rate may be a useless metric
Machine learning in R. Homeschooling + working.
The evolution of the American Census
Slack guidelines. Rice measurements. Hadley Wickham interview.
How Netflix measures product success
R packages for exploratory analysis. Building data science infrastructure. Making model diagrams.
Health and safety first. Bracing for economic impact second.
Weathering economic headwinds. Building DS infrastructure. Visualizing different things about COVID-19.
How the coronavirus might impact hospitals
COVID-19 tracking project. Data inclusivity. Medical lit primer.
Data science at Stitch Fix
Responsible coronavirus viz. Going remote. Exponential growth.
Data centers are the new oil
Twitter for R programmers. Big NLP database. DS research challenge areas.
Spotify’s data discovery platform
Grubhub’s model deployment tools. Deep learning book series. Gender identify data.
Managing Gitlab's data team
How AI businesses are different from software. Data-driven weather forecasting. Recommender model metrics.
The data engineering hierarchy of needs
Understanding uncertainty. Bokeh tutorial. Everyday dev tools.
Google’s on a ML publishing streak
10x data scientists? No thanks. A Timsort history. Flow fields.
Avoiding the coronavirus on your flight
Farewell, OLAP cube. Data training for the underserved. Reinforcement learning curriculum.
Data science consulting: a retrospective
Wayfair product ranking. Misconceptions about names. A first grade deep learning model.
Trailing vs. leading commas
Data engineering observability. Global power lines. Code skills check-in.
Data viz goes mainstream
4 helpful distribution charts. A data project checklist. Interactive tools for ML.
A decade of differences
Machine learning for the real world. A fun SQL problem. An R cookbook.
State of JS. Getting help in R. Keeping it simple.
Instagram’s recommender system
ACLU’s data transformations. Board games + data viz. User-Agent history.
Biased algorithms vs biased people
Deep learning in production. AI text adventures. Guido Van Rossum.
Machine learning systems design. Calculating customer types. Democratic primary.
Algorithms in the courtroom
Film flowers. A new R color palette. Machine learning engineering.
The rise of the analytics engineer
Working in football analytics. “Biased data.” Churn correlations.
Finding consulting clients
A funny dataset. A cleaning package. A new way to measure intelligence.
Character encodings. Detecting audio deepfakes. DS archetypes.
Better buffet lines
Boring technology. Metrics problems. Racially-biased medical algorithms.
2019 coding salaries
PyTorch vs TensorFlow. Country goodness. A neural net rabbit hole.
Introducing Helix. Time series databases. Fall foliage.
How America moves its homeless
1-line data exploration. Survival analysis. Handling dates.
Neural network design. Chart taxonomies. Analyzing Amazon purchases.
Flunking the fourth down
Red flags in interviews. Data integrity. Fashion algorithm.
Lyft’s hyper-accurate maps. Neural net DM. Selling data science.
The care and feeding of data scientists
Data sculptures. Load testing. Abstractions.
Credit card conundrums
DS salaries. UX of data. Stats 101.
Lyft’s query engine
Speedy ML. Fitness app data. NPS overnight stays.
Exploring Bob Ross paintings
Mastering Shiny. DeepMind’s losses. Music trends.
Hackers gonna hack
More time, please. Pandas tricks. Animating data viz.
Stitch Fix’s centralized experimental platform
Machine learning... learning. Behavior funnels. DS team models.
Lyft’s self-driving dataset
Vision science. R-Ladies. “De-identification.
Team names. Podcast structures. ML best practices.
Free NLP course. Tidy log odds ratio. Colorspace.
2019 data & AI landscape
Practical psychology for DS. Filling in missing music. Deploying models and microservices.
Why you swipe right
BS AI industrial complex. Marketing automation at Lyft. Word2vec.
Python is weird
Hadoop’s failure. GANs everywhere. AI adoption.
Deepfake propaganda isn't a thing
Neural nets name cats. Uncertainty in viz. Misunderstood data engineers.
Python's Caduceus syndrome
Fashion deepfakes. Research quality data. Instagram analysis.
Predictive grocery bills
Career advice. Liverpool's analytics. Type stable estimation.
Advice from a Lyft data scientist
Fullstack D3. ML product management lessons. Matrices as tensor network diagrams.
Stripe's ML infrastructure
Altair. Railyard. profvis.
Make your own data
AI death metal
Fooling AI surveillance. Skittles math. Data viz on mobile and desktop.
Addicted to models
DataCamp reflections. dplyr filter. Deadline statistics.
How Uber scales customer support
Machine learning job hunt. Open GAN questions. The English block on programming.
Unintended consequences. Partnership culture. Active learning data labeling.
Crimes against data viz
Data viz freelancing. Unpacking adversarial examples. Null reminder.
Automated Instagram influencers
Escaping Excel hell. P-value put down. Harmonizing with Bach.
DS vs ML vs AI
Advice for new data scientists. R surprises. Specialists vs generalists.
The AI Diet
The art of analytical persuasion. Deep learning & mind reading. Time-to-event data.
A cocktail's nearest neighbors
AI for rats. SQL vs Python for pipelines. NYT's data editor.
You need a query library
Visual search engines. Face editing GANs. Tidy Tuesday.
SQL is a super power
Data versioning. Deep learning limitations. Successful career transitions.
Farewell, deep learning?
Podcasts. Data viz criticism. Health risk scores.
Deepfakes blink first
ML project management. Analytics engineer. Minimally sufficient Pandas.
30 minutes to answers
Data science jobs up 29%. Rstudio::conf recap. Curriculum roadmap.
Moneyball for college
Why analytics initiatives fail. Model representation in R. Bye, mid-range shot.
Data engineer required?
Uber's new AI. Location data for sale. Data roadmapping.
Tech talk prep
2018 AI trends. Gender bias in Google Images. The power of great analysts.
Mo models, mo problems
190+ R-stats tasks. KDEs. Selection bias.
The Netflix data war
The saddest Christmas song. Data science vs engineering. Mongo → Postgres.
The grossest stadium food
Next-gen GANs. Tidy tutorial. AI art gallery.
How Instacart predicts grocery item availability
5 soft skills analysts need. State of deep learning. AI detected fields and crops.
The evolution of chess
Airflow on your Macbook. Modeling time-lagged conversion rates. Google Earth open data.
Craft breweries are common ground
Public opinion on algorithms. Obscure Tidyverse packages. Gov't data troves.
Is this actually AI?
Data dictionaries. New McKinsey AI survey. Tidy Tuesday.
Analysis over dashboards
Analytics-DevOps harmony. Tidyeval tutorial. Client engagement.
3 million election ads. Other data science tools. Montezuma’s Revenge.
AI art “thieves”
Satisfaction in ML careers. Making your product freemium. Relearning code.
Seeing isn't believing
A summer at Airbnb. Deepfakes look too real. Analysts on production.
Tricks for estimating uncertainty
Amazon's secret AI recruiting tool. Data PMs. Altair tutorial.
Artwork personalization at Netflix
Chromebook data science. Shaky CDC data. An NLP history lesson.
Build your own deep learning computer
Citizen data scientists. ML for coders. Connections across America.
Anatomy of an AI system. Reproducing ML projects. 5 public datasets to explore.
Retracing your steps in machine learning
Strata slides. A DS ethics checklist. Life expectancy in your neighborhood.
Who is Anonymous?
PhD considerations. SQL vs Python. Machine translators.
China reigns with data
California's wild fires. Integrating Ml into a product. Twitter toxicity.
The patriarchy of pockets
SQL queries for Salesforce. The beauty of annotations. Capturing data evolution.
The hazards of A/B testing
BigQuery table clusters. Unpacking an NLP Twitter thread. First days on the job.
Know your blindspot
3 million troll tweets. The Holy Grail of email. Partitioning variation.
IMDb analysis. Differentiatiable image parameterizations. ACL 2018 highlights.
Reinforcement learning's fundamental flaw
W. E. B. Du Bois' data viz. Machine learning glossary. Speedier R work.
Doing good data science
What do ML practitioners do? Ditching microservices. Feature-wise transformations.
Apple Maps is reborn
What 85-year-olds are up to. Red flags in interviews. 12 ggplot2 extensions.
LeBron's next pick
2018 data viz survey. Data engineering frameworks. ML papers with code.
The best Mario Kart character
Constrained optimization. Census oddities. Bias-variance tradeoff.
Predicting the World Cup
ML's future is tiny. One year as a data scientist. Problems with gender classification.
Remote vs non-remote workers
Shazam, for Congress. The future of data engineering. Trustworthy data analysis.
Is UTC enough?
Better training data. Rethinking academic data sharing. Volcanic history.
An abundance of Ns
purrr tutorial. LaCroix color palettes. The challenges of Smart Compose.
The structure of standup
Strategies for optimizing Python code. The NYC subway crisis. Russian Facebook ads.
Attracting top notch candidates. Data violence. Linear vs log scale.
The taxonomy of food
Grubhub's seemingly impossible data problem. Qualitative before quantitative. R package dependencies.
Mode Studio. A Shiny app for Fido. Streaming 100 billion analytics events.
The future is modular
Why data scientists should take a hippocratic oath. Machine learning at Conde Nast. New viz tools.
CNN vs MSNBC vs Fox
Odes to notebooks. Overcoming objections. Lumpers and splitters.
The evolution of Stephen Curry
Academia → industry. Lessons from video games. TensorFlow sans setup.
You've got one (million) shot(s)
A massive NCAA data set. GANs + art. The benefits of blameless postmortems.
Stack Overflow's developer survey. Docker for deep learning. Data-driven unit testing.
How neural networks decide
SQL → Pandas. Prophet in Mode. Visualizing outliers.
300 years of data viz
Love for the star schema. 8 in-app analytics examples. “Wall time” semantics.
How long's the wait?
Down with pipeline debt. Malicious use of AI.
All you knead is love
Pythonic cookies. A data viz engineer definition. Manifesto for data practices.
The Olympics. ML models you haven't built. Visualizing missing data.
Why recommendation engines fail
Awesome in-app analytics. Another data privacy mishap. DJ Patil rallies the troops.
The Oscars game plan
Window functions in Python & SQL. The future of pandas. Free resource roundup.
Imposter no longer
Data engineering for dummies. The mortality rate of JS frameworks. A new DS podcast.
Follow the tea trails
Graphics reporter Q&A. Automating front-end. Job hunting post mortem.
Where athletes come from
Selecting a cloud provider. Academia → industry. Early-stage analytics.
No free lunch
Junior DS roles. A literal gamechanger. ML technical debt.
Google Maps' Moat
The next Bechdel Test. PyCon proposal myths. A re:Invent recap.
Now in 3D!
Apache Airflow, explained. The 3rd dimension of customer success analysis. Molecule's custom reports.
Postgres is hip again
U.S. AI threatened. NIPS highlights. Median pitfalls.
Hadoop or laptop?
Netflix's A/B test alternative. Predicting palliative care. Building a deep learning library.
A measure of fairness
Ethics in practice. Data meta-metrics. Numerical optimization.
Tracking your Thanksgiving
Crossword heatmaps. Generative music. Tips for building a diverse data team.
Trophy data scientist
The 3-degree world. Causal inference. Changepoint analysis.
Halloween episodes. Ethical responsibility. Word cloud designs.
Streaming Spotify data
4 data mistakes startups make. Interviewing data scientists. A massive font database.
Trust the process
NBA analytics. Power calculations. State of data journalism.
A dirty dozen
12 A/B test pitfalls to avoid. Taxis vs cabs vs the subway. Evaluating ETL tools.
The Nate Silver Effect
zulily's data platform. R for journalists. The NYC job search.
Debunking studio exec claims. An ETL company's stack. What closes deals.
Problems with probability
Accelerating GeoPandas. New R community. Making analytics meaningful.
Data security for data scientists
Finding the best data jobs. Legos + text mining. Scalable machine learning.
Troubleshooting neural networks
10x data scientists. 30 years of hurricanes. Communicating uncertainty.
Improving the Zestimate
Language gaps. Packaging metrics. A Python cheat sheet.
The Data Trust Gap
Foursquare's location intelligence. Giving your first data science talk. 10 Python mistakes.
Optimizing for Burning Man. Choosing an ETL tool. Scaling with Python.
Cargo cult data science. Millions of Intercom messages. Query optimization.
Facebook's AI factory
Predicting LTV at Airbnb. Technical debt in ML. What's difficult about histograms.
Machine learning at Apple
Gender representation in comics. Data systems. Designing enterprise tables.
Graphing Jane Austen's genius
Joy plots. How to spot a misleading graph. Marrying UX & ML.
Rise of the data PM. Augmented reality viz. New NYC boroughs.
Optimizing Reddit submissions. R at Microsoft. Blogging about data.
A tale of two axes
Millions of doodles. 2 years at Stack Overflow. Coding on the go.
3 stages of data infrastructure. 29 common Python errors. 200,000 Uber and Lyft trips.
Big data B.S.
Root cause analyses. How histograms work. Analytics at Athos.
The MLB's new metric. How to hire a product analyst. The Paris Agreement.
Mr. Rogers' rainbow
Airbnb's Data University. 30 GBs of federal payroll records. The top DS software.
The Emoji States of America
Big news from Mode. The Hitchhiker's Guide to d3.js. Detecting overspend in AWS.
Counting the hours
Duolingo's language learning model. Etsy's development process. Instacart's strategy for building DS teams.
Airbnb's North Star
Winning marital arguments with R. 3 million Instacart orders. Dashboards that deliver.
100 billion events
Spotify's event delivery system. Craft beers and Python. Data viz vs UI.
Architecture of Giants
Machine learning flash cards. Teaching SQL. Analytics trends in 2017.
The Stats of the Furious
Statistics in D3. Proving yourself without a degree. More on interactive viz.
I saw the sine
Avoiding analytic rabbit holes. The Data Wheel of Death. Rebuilding an analytics stack.
Winning at Scrabble
ML for product managers. Analytics for startup founders. Scrabble analyses.
Group-by from scratch. Corporate data viz. Test-driving Prophet.
Perl is dead
Switching programming languages. Data hackathons. Is interactive viz done for?
Open source burnout
A data GIF tutorial. DS on the Silicon Beach. Blind date data.
Testing time series
Hiring a data scientist. The future of Airflow. Advice for switching careers.
The Zero Bug
Predicting earthquake preparedness, partisan conflict, and feature engineering.
Online DS courses, ranked. Critical data literacy. Unlearning descriptive statistics.
Remembering Hans Rosling
Spotting visualization lies. Data humanism. Encoding categorical values.
Finding fake news
Mode's stance on Trump. ML at Fitbit. The cleanest NYC restaurants.
The hottest year yet—again
Data science at Stitchfix. ML videos. A data engineer's manifesto.
How stats lost their power
Redefining “AI.” Behind-the-scenes of sports analytics. Building a master data dictionary.
A freelancer's tale
Uber Movement. Q&A w/ Monica Rogati. Visual vocabulary.
Sack the coach
The NFL and causal inference. Generating poetry with Python. Classifiers from scratch.
The great TV divide
Mid-career pivots. TV fandoms. Rationality + empathy.
Bringing down the Empire
Star Wars casualties. The state of the DS job market. CAC calculations.
#DataRefuge. A chat with DJ Patil. Analyzing Google trends.
Analyze the rainbow
Skittles debates. Time series analysis in Python. Ditching vanity metrics.
Look ma, no polls!
A data detective story. BitTorrent for professors. Seasonality in search engines.
The good ol' days
Rebuilding trust in analytics, data limitations, and a text analysis tutorial.
Data wrangling Westworld
Data skills we all need, election post mortems, and runner routes.
The non-election issue
UFO sightings data. 415 viz tools. The science of unpredictability in... science.
When charts lie
Why data projects fail. An AI speechwriter. The end of baseball's analytics war.
Double the Trump
Flash forecasting. The father of soccer analytics. A new viz technique.
The David Spade Index
The problem with North Star metrics, the secret to designing smart products, and the popular vote.
Data in the deep fryer
The impact of outliers, marathon performance, and why machine learning is like deep frying.
Studying The Simpsons
Nobel Prize winners, your typical farmers market, and The Simpsons side characters.
Data movie magic
The year's best data visualizations, fact checking the debate, and movie magic with data viz.
Polling the pollsters
Gender roles in Hollywood, stats for soccer fans, and four results from one election poll.
Summary analysis, creativity in data viz, and the income increase.
The state of data engineering
Digital economists, swing states, the art of asking good questions.
What is Bayesian, really?
The pros and cons of urban cycling, rebuilding a Graphics team, and the joys of dot plots.
In English, please
One color palette generator, 8 Python data cleaning libraries, and the fastest men in the world.
End the language war
Visualizing clickbait, counting conundrums, and the problem with the Rio pool.
Data journalists go for gold
Trump tweets, Olympic data viz, and tips for designing better tables.
A Star Trek network viz, ethics for algorithms, and the Olympics.
Lies and statistics
Data viz developments, dodgy statistics, and genomics.
Amazon reviews, Bayesian thinking explained visually, and dashboard design.
Half a decade of drought
Pop music genealogy, FiveThirtyEight's R workflow, and a series of stunning drought maps.
Icarus, Oedipus, Cinderella
Data mining story arcs, theories of everything, and the history of the infographic.
Brunch so hard
Feature engineering, cartogram challenges, and an analysis Leslie Knope would love.
Plus: Mode now supports Plotly, data science portfolios, and fantasy football.
The 'Hamilton' algorithm
Escaping Excel hell, real-time dashboards, and the Data Journalism Awards.
Reddit's favorite field of science
Pie chart research, a Python cheat sheet, and machine learning for sales.
The influence of 'In da Club'
50 years of pop music, hybrid intelligence, and HBR data viz advice.
Following in Apple’s footsteps
Tighter data security, Foursquare + Uber, and data anonymization best practices.
Calculus not required (plus big news from Mode!)
News on the Python front, 24 charting tools, and SF rental prices.
A tale of two types of journalism
Sales data, pandas video tutorials, and data science in healthcare.
Beyond 'The Touchscreen Generation'
Kalman filters, pandas tutorials, and why newsrooms should own their data.
Foursquare the prophet
The power of proprietary data, pirated papers, and a glorious data viz catalog.
Googling 'Game of Thrones'
12 data science methods and 1 big HBO show.
Kobe's last shot
Thumbtack's data stack, storytelling at Jawbone, and 15 data viz interviews.
Big Brother grew wings
FBI spy planes, measuring MRR, and the Hollywood gender gap.
Mathematics is coming
Game of Thrones, scaling the data science org, and conversion optimization.
Pity the pie chart
Lift analysis, genocidal chatbots, and the plight of pie charts.
Moneyball for book publishers, CAC, and the engineer-analyst relationship.
Back-end analytics, help center metrics, and predicting police misconduct.
Pointers from Steph
Shattering NBA records, 2 million chess games, and statistically significant growth hacking.
Break out the bubbly
Punctuation in code, PM employee onboarding advice, and practical data science skills.
Do the math
If Facebook were a pollster, BuzzFeed analytics, and the virtues of keeping things simple.
A solar system of 'Tainted Love'
Interstellar cover songs, 10 TED talks, and the presidential primary.
Mind the map gap
Central Africa's dearth of data, alternatives to open data portals, and data viz empathy.
A new kind of parasite
Flint failings, research parasites, and Disney princess linguistics.
13 years after Moneyball
Sabermetrics, DAU, and holiday shopper retention.
Squirrels gone wild
Edtech analytics, Python prep, and Powerball.
Hamlet's social network
Missing ordinals, football analytics, deep-learning chips, and more.
A decade in the life
'Star Wars,' random projection, inviting dissent, and Nick Felton's final report.
Getting closer to 'Her'
A machine intelligence progress report, mesmerizing viz, insightful data science talks, and delivery analytics.
The fashionable side of data science
Best practices, Google's effect on the 2016 election, climate change, p-values, and data-related stocking stuffers.
Smiles, agriculture, Airbnb's data release, and more.