Insights in Software, Data, and AI

Feature stores explained: when you need one and how to build it

When you need a feature store for ML and how to build one, explained.

August 18, 2026

10 min of reading

Illustration of a document AI pipeline turning a stack of scanned pages into a structured data table

Artificial Intelligence

Document AI: turn scanned PDFs into structured data

How a document AI pipeline turns scanned PDFs into structured, trusted data.

August 17, 2026

10 min of reading

Line-art illustration of a document being read by a fine-tuned OCR model, with a gear and neural network, on the BIX Tech blue background

Artificial Intelligence

Data Science

Fine-tuning an OCR model in 2026: an updated step-by-step guide

Step-by-step guide to fine-tuning an OCR model in 2026.

August 17, 2026

9 min of reading

Illustration of a Power BI analytics report window with a usage chart and a magnifying glass exporting data to a CSV file, showing how to extract Power BI usage data with no code

Extracting Power BI usage data without the REST API: a no-code guide

Extract Power BI usage data with no code, without the REST API.

August 16, 2026

9 min of reading

Illustration of SQL window functions: a sliding window framing rows of a data table, with LAG and LEAD arrows and a moving-average line chart

Window functions in SQL: advanced patterns every analyst should know

Advanced SQL window functions every analyst should know

August 16, 2026

9 min of reading

Power BI row-level security at scale: a user's identity filters a table so only permitted rows come back

Power BI RLS at scale: workspaces, pipelines and row-level security

Power BI RLS step by step: workspace roles, deployment pipelines and dynamic row-level security.

August 14, 2026

12 min of reading

Line-art illustration of a slider, a rising SKU staircase, a utilization gauge and a cloud capacity stack, representing how to choose and right-size a Fabric capacity

Power BI Premium vs Fabric: how to choose and right-size your capacity

Power BI Premium vs Fabric: SKU mapping, cost per scenario and sizing

August 12, 2026

13 min of reading

Line-art illustration of a model package, a serving endpoint, a drift monitoring panel and a retraining loop in a circle, representing an MLOps reference architecture

Data Science

MLOps in 2026: reference architecture for model deployment

An MLOps reference architecture for model deployment and monitoring.

August 10, 2026

13 min of reading

Line-art illustration of a speedometer, a bar chart, a stopwatch and a database cylinder linked in a clockwise cycle, representing Power BI performance tuning