Best of Data Science — August 2025

1. 1
   Article

   Ars Technica·39w

   College student’s “time travel” AI experiment accidentally outputs real 1834 history

   A computer science student created TimeCapsuleLLM, an AI language model trained exclusively on Victorian-era London texts from 1800-1875. When prompted with "It was the year of our Lord 1834," the model unexpectedly generated text referencing real historical protests and Lord Palmerston's actions from that exact year. The student discovered through fact-checking that these were actual historical events, demonstrating how AI models trained on period texts can inadvertently capture and reproduce authentic historical information. This project is part of a growing field of Historical Large Language Models (HLLMs) that aim to recreate linguistic patterns and knowledge frameworks from past eras.

   56

   16
2. 2
   Video

   Tech With Tim·42w

   Learn Pandas in 30 Minutes - Python Pandas Tutorial

   A comprehensive beginner tutorial covering pandas fundamentals including dataframe creation, data loading from CSV files, basic operations like head/tail/info, column and row indexing with iloc/loc, filtering data with conditions, updating and deleting entries, data cleaning methods, and basic analysis functions like groupby and value_counts. The tutorial demonstrates both regular Python files and Jupyter notebooks for data manipulation workflows.

   25

   1
3. 3
   Video

   Tech With Tim·39w

   Python is Changing – Here’s What’s Coming

   The Python Developer Survey 2024 reveals key trends shaping Python's future. FastAPI is overtaking Flask and Django for API development, while data science usage has grown to 51% of developers. Popular tools include pandas and numpy for data analysis, PyTorch gaining ground over TensorFlow for ML, and new package managers like UV showing rapid adoption. Most Python developers have less than 2 years of professional experience, making expertise valuable. The survey highlights Python's continued dominance in web development, data science, and machine learning, with emerging tools like Streamlit for dashboards and Pydantic for validation becoming standard.

   15

   1
4. 4
   Article

   freeCodeCamp·41w

   Learn Time Series Forecasting in Python

   A comprehensive Python course covering time series forecasting fundamentals, including data decomposition, ARIMA models, cross-validation techniques, and evaluation metrics. The course teaches how to build baseline models, incorporate exogenous features, generate prediction intervals, and apply practical forecasting techniques for predicting future trends.

   10