Let's Data Science · Oct 23, 2026 · 3:08 PM UTC

Let's Data Science

A security patch can stop the exploit you tested and still leave the flaw open. In an LDS Exclusive, @Follow_our_Echo security researcher Yuval Saggie explains what a fix must prove before you trust it, sometimes before a CVE number exists. letsdatascience.com/news/ech…

ALT LDS Exclusive interview cover: Before a CVE, verify the fix. Yuval Saggie, Security Researcher, Echo.

Let's Data Science · Oct 23, 2026 · 3:02 PM UTC

Let's Data Science@letsdatascience3h

In Nimble's test, two web agents had the same Python data. One stopped at 153 of 375 rows after 13 minutes. The other wrote code and got all 375 in 90 seconds. In an LDS Exclusive, @nimble_search's Yuval Bertocchi explains when agents should run code. letsdatascience.com/news/nim…

ALT LDS Exclusive interview cover: When web agents should run code. Yuval Bertocchi, Director of Engineering, Nimble.

Let's Data Science · Oct 22, 2026 · 2:18 PM UTC

Let's Data Science@letsdatascienceOct 22

How it works: 1) AI News Quiz, daily 2) AI Word, daily 3) AI Mini crossword, daily 4) Weekly AI Crossword, Mondays Up to 100 points per edition, all added up. Free account needed; guest play doesn't count. First attempt only, no hints. Rules: letsdatascience.com/news/gam…

ALT AI News Quiz card: five questions from this week in AI. Daily, free to play at letsdatascience.com/news/games/quiz.

ALT AI Word card: one AI word, six tries, with a clue from the news. Daily, free to play.

ALT AI Mini card: a tiny crossword with today's AI news inside. Daily, free to play.

ALT Weekly AI Crossword card: a bigger grid built from the week in AI, new every Monday. Free to play.

Let's Data Science · Oct 22, 2026 · 2:18 PM UTC

Let's Data Science@letsdatascienceOct 22

The prize is Claude Max. The catch: no Claude, or any AI, while you play. Four free AI news games, one October leaderboard. Top player wins a month of Claude Max 5x (US$100 value). Under 10 days left. Leader: 580 points. Over 3,000 still in play. letsds.com/news/games

🤖 Made with AI

LDS News Games monthly challenge: play four AI news games for a chance to win one month of Claude Max 5x, a US$100 subscription value. Free to play. Rules apply. Funded by LDS, not sponsored by Anthropic.

ALT LDS News Games monthly challenge: play four AI news games for a chance to win one month of Claude Max 5x, a US$100 subscription value. Free to play. Rules apply. Funded by LDS, not sponsored by Anthropic.

Let's Data Science · Oct 21, 2026 · 8:06 PM UTC

Let's Data Science@letsdatascienceOct 21

95% accuracy, zero positives found. In this synthetic 100-case example, always predicting negative misses all five positives. Put the confusion matrix in your ML portfolio. letsdatascience.com/learn/pa… #DataScience

ALT Synthetic ML portfolio example with 100 cases: 95 negatives and five positives. A baseline predicts negative for every case. The confusion matrix has 95 true negatives, zero false positives, five false negatives and zero true positives. Accuracy is 95/100, or 95%; positive recall is 0/5, or 0%. Show class counts and the confusion matrix alongside the score. Conceptual illustration, not a product result. Official LDS logo, Let's Data Science and letsdatascience.com appear in the footer.

Let's Data Science · Oct 21, 2026 · 3:07 PM UTC

Let's Data Science@letsdatascienceOct 21

"Active in September" can mean any service during the month or still active on September 30. An AI analyst should resolve that choice before writing SQL. letsdatascience.com/blog/tex… #DataScience

ALT Text-to-SQL flow: look up approved definitions for active customers in September. Service on any day in the month and still active on September 30 are different definitions. If both fit, ask the user to choose, then generate SQL with the selected definition. Validation and permissions still follow. Includes the official LDS logo, Let's Data Science and letsdatascience.com.

Let's Data Science · Oct 20, 2026 · 8:07 PM UTC

Let's Data Science@letsdatascienceOct 20

One $100 order, two line items, a $200 sum. The join repeated an order-level measure at item grain. Add a before/after total check to your analytics engineering portfolio. letsdatascience.com/learn/pa… #DataScience

Comparison of one order row with two joined item rows. Order 101 totals $100. Joining to items A and B repeats $100 in both rows, so a naive sum gives $200. Calculate order revenue at one row per order. The illustrative graphic includes the official LDS symbol, Let's Data Science and letsdatascience.com.

ALT Comparison of one order row with two joined item rows. Order 101 totals $100. Joining to items A and B repeats $100 in both rows, so a naive sum gives $200. Calculate order revenue at one row per order. The illustrative graphic includes the official LDS symbol, Let's Data Science and letsdatascience.com.

Let's Data Science · Oct 20, 2026 · 3:07 PM UTC

Let's Data Science@letsdatascienceOct 20

model.eval() does not turn off autograd. torch.no_grad() does not turn off nn.Dropout. Test both controls in your ML serving portfolio. letsdatascience.com/learn/pa… #DataScience

Decision map for a PyTorch forward pass. Two independent checks lead to model.eval(), which disables nn.Dropout but does not disable autograd, and torch.no_grad(), which skips gradient recording but does not set eval mode. The branches join at a three-line example using both before y = model(x). A portfolio takeaway asks for separate tests of layer mode and gradient tracking. The official LDS symbol, Let's Data Science and letsdatascience.com appear in the footer.

ALT Decision map for a PyTorch forward pass. Two independent checks lead to model.eval(), which disables nn.Dropout but does not disable autograd, and torch.no_grad(), which skips gradient recording but does not set eval mode. The branches join at a three-line example using both before y = model(x). A portfolio takeaway asks for separate tests of layer mode and gradient tracking. The official LDS symbol, Let's Data Science and letsdatascience.com appear in the footer.

Let's Data Science · Oct 19, 2026 · 8:11 PM UTC

Let's Data Science@letsdatascienceOct 19

Your test label stayed hidden. Its feature value still leaked into the training mean. Split first, fit preprocessing on training, then transform test with the same fitted rule. letsdatascience.com/learn/ml… #DataScience

ALT Two-panel comic illustrating ML preprocessing leakage. A practitioner centers all rows before splitting: training values 0 and 2 plus test value 100 give mean 34. A colleague separates the test row and fits mean 1 on training only. Subtracting that mean gives training values -1 and 1, and test value 99. The takeaway is to fit on training and apply the same transform to test. Official LDS logo, Let's Data Science and letsdatascience.com appear below.

Let's Data Science · Oct 19, 2026 · 3:07 PM UTC

Let's Data Science@letsdatascienceOct 19

Trace a residual update: [2, -1] + [0.5, 0.25] = [2.5, -0.75]. In this Pre-LN branch, only the update path is normalized. Set the update to zero to check the skip path. letsdatascience.com/learn/bu… #DataScience

One pre-normalized transformer residual branch, using illustrative values. Input x=[2,-1] splits: a teal skip path keeps x; an orange path runs through LayerNorm and a sublayer to give update=[0.5,0.25]. Addition gives y=[2.5,-0.75]. The equation is y=x+Sublayer(LayerNorm(x)). If the update is zero, y=x. This is one branch, not a full block. The official LDS symbol, Let's Data Science and letsdatascience.com appear below.

ALT One pre-normalized transformer residual branch, using illustrative values. Input x=[2,-1] splits: a teal skip path keeps x; an orange path runs through LayerNorm and a sublayer to give update=[0.5,0.25]. Addition gives y=[2.5,-0.75]. The equation is y=x+Sublayer(LayerNorm(x)). If the update is zero, y=x. This is one branch, not a full block. The official LDS symbol, Let's Data Science and letsdatascience.com appear below.