AI for the people who work the data

Tested, hands-on guidance on using AI across the analytics workflow, from text-to-SQL to dashboard design, written for analysts and data scientists rather than executives.

ADVERTISEMENT

Free tools

All tools →

Every tool runs entirely in your browser. Nothing you type is uploaded or stored.

Free tool

AI API Cost Calculator

Work out what an AI model actually costs per month from your token usage, and compare the major models side by side.

Free tool

Token Estimator

Paste any text to estimate how many tokens it uses, and see what that text would cost to send to each major model.

Free tool

Subscription vs API: Which Is Cheaper?

Compare a flat monthly chat subscription against the equivalent API usage and find the break-even point where one overtakes the other.

Free tool

AI Tool Picker

Four questions about your task, budget and experience — then a recommendation and the guides to go with it.

Browse by topic

All guides →
3 guides

AI Foundations for Analysts

The mental models a data professional needs to use AI well, from how language models read a schema to where embeddings beat classic feature engineering.

3 guides

Talk to Your Data

Setting up and trusting conversational and text-to-SQL assistants that let you and your stakeholders query a warehouse in plain language.

3 guides

Insight and Narrative

Using AI to turn model output and query results into memos, documentation, and summaries that survive stakeholder scrutiny.

3 guides

Dashboards and Data Viz

Applying AI to the visual side of analytics, from generating chart specs to catching accessibility and layout problems before a dashboard ships.

3 guides

Automation and Pipelines

The automation need: prompting for correct SQL and pandas, and wiring LLM steps into pipelines that clean, triage, and self-heal.

3 guides

Data Strategy and Career

The money and career side: costing AI-in-the-loop workflows, winning budget, and understanding how these tools reshape the analyst ladder.

ADVERTISEMENT

Start here

All guides →

How Large Language Models Actually Parse a SQL Schema, and Where They Get It Wrong

A model does not read your database the way you do. Understanding what it actually sees explains most of its confident, wrong joins.

Ravi Chandrasekaran, a former data science lead turned AI educator · 9 min read

Setting Up a Text-to-SQL Assistant Against Your Warehouse Without Exposing Raw Tables

A practical architecture for letting analysts ask questions in English while the model never sees your raw schema, PII, or ungoverned joins.

Grace Okafor, an analytics engineer specializing in natural-language querying · 9 min read

Estimating the Real Cost of an LLM-in-the-Loop Analytics Workflow at Scale

Token prices are the smallest line item. Here is a cost model that accounts for retries, context bloat, human review, and the warehouse bill underneath it.

Yara Haddad, a data strategy consultant · 9 min read

Building a Python Pipeline That Self-Heals Broken Column Mappings With an LLM Step

How to catch renamed and reordered upstream columns automatically without letting a language model silently corrupt your schema.

Dmitri Volkov, a senior analytics automation engineer · 9 min read

Why Conversational BI Tools Hallucinate Metrics, and the Guardrails That Actually Reduce It

The failure isn't bad SQL syntax — it's confident, wrong numbers that look exactly like right ones. Here's where they come from and how to catch them.

Grace Okafor, an analytics engineer specializing in natural-language querying · 8 min read

Critiquing an Executive Dashboard With AI Before It Ships

A vision model won't tell you if your numbers are right, but it will catch the layout problems you stopped seeing three revisions ago.

Hana Lindqvist, a data visualization designer · 8 min read

Why this site exists

Most AI coverage is written for everyone or for no one.

We write for the person staring at a broken join at 6pm, not the keynote audience. Every recommendation here comes from someone who has shipped a dashboard, defended a number, or debugged a pipeline that an AI tool confidently got wrong.

Read more about who we are and our editorial standards.