Posts

Intro It’s a short yet handy guide on how to analyze time series data in a few clicks using modern Python libraries. Tech stack for this guide consists of duckdb and pandas packages for data processing Facebook’s prophet for data exploration and forecasting plotly for data visualization Code import duckdb as db import pandas as pd import numpy as np orders = pd.read_csv('orders_data.csv') orders.sample() ...

How MD5-based traffic splitting with salts and a layer system lets you run dozens of orthogonal experiments without conflicts

Classical and Poisson bootstrap as non-parametric tools for normalizing skewed metrics and enabling valid T-tests on small samples

How aggregating observations into buckets normalises skewed metrics, reduces variance, and cuts computation — with a probability proof for choosing the right bucket count