| 🔗 Note: if you don’t see links in this email, it’s also available on the web: dataelixir.com/archive/issue-582.html |
| | |
| | | AI tools have moved beyond chat, and choosing one now means deciding what kind of work you want it to do, what it has access to, and how closely you want to supervise it. In this guide, Ethan Mollick walks through the increasingly confusing options and offers a practical way to start experimenting with them.
Ethan Mollick |
| | |
| | | Sometimes the challenge with time series isn’t your forecasting model. It’s that the future simply doesn’t resemble the past closely enough. This post explains why better algorithms can’t manufacture signal, and shows why understanding where the data comes from often matters more than adding model complexity.
Suzanna Sia |
| | |
| | This side-by-side reference compares common data manipulation operations across data.table, base R, dplyr, polars, and pandas. It covers filtering, grouping, transformations, joins, reshaping, and more, with equivalent code for each framework.
Vincent Arel-Bundock |
| | |
| | Adding an index is often the first fix for a slow query, but that will only get you so far. This production-tested Postgres guide builds a much better mental model, covering query plans, locks, vacuuming, connection pooling, migrations, and the operational problems that appear as databases grow.
Hatchet | Alexander Belanger |
| | |
| | In this intermediate level course from Coursera, learn how to conduct statistical analysis, critique data visualization, apply machine learning, analyze social network connectivity, and much more. Enroll for free.
*Message from this week's partner, Coursera |
| | |
| | | | A strong data science portfolio doesn’t need a groundbreaking idea. It needs to show how you choose problems, evaluate results, explain limitations, and communicate what you learned. This guide walks through how to build one.
Ken Reid |
| | |
| | | When AI makes code easier to produce, proving that it works becomes the hard part. This paper uses eight scientific computing projects to show why specifications, tests, and expert judgment matter more than ever.
OpenAI Research |
| | |
| | Supply chains can amplify shocks even when everyone has perfect information. This paper explains why, then turns the idea into an interactive model where you can disrupt major oil chokepoints and watch shortages spread across the global network. The main link opens the visualization and there's a link to the paper at the top of the page.
Elioth Sanabria |
| | |
| | | Synthesizers are basically tiny math machines for making sound. This interactive explainer starts with sine waves and harmonics, then builds up to oscillators, filters, envelopes, and a playable browser synth.
Shri Khalpada |
| | |
| | |
|
