INTELLIGENCE
Guides
How to Get Started with Count for Data Analysis
Count is a collaborative agentic analytics platform where teams can clean, model, analyze, and visualize data in one canvas using SQL, Python, and natural language. Users connect a warehouse or upload a CSV, then ask questions through a built-in AI agent backed by models from Anthropic, OpenAI, and Google. Every query and chart the agent produces is fully editable, making Count usable by both non-technical analysts and data engineers working side by side.
How to Set Up DuckDB for Local Data Analysis
DuckDB is an in-process analytical database that runs SQL queries on CSV, Parquet, and JSON files directly from your laptop, with no server to configure. Version 1.5.2, released April 13, 2026, adds DuckLake v1.0 format support and performance improvements. You install it with a single pip command and run the first SQL query on a CSV file in seconds. No cloud account, no Docker, no database administrator required.
How to Set Up Copilot in Power BI
Copilot in Power BI uses generative AI to create report pages, summarize data, and generate DAX calculations from plain-English prompts. Enabling it requires a Microsoft Fabric capacity of F64 or higher, or Premium Per User licensing tied to a capacity-backed workspace, plus a Fabric administrator toggling the feature on at the tenant level. Once configured, analysts can generate a full multi-visual report page from a single sentence.
How to Build an AI Data Analyst Chatbot with n8n
n8n is an open-source workflow automation platform with over 400 integrations. Its AI Agent node, combined with data retrieval tools and a Calculator node, lets you build a chatbot that connects to a live spreadsheet or database and answers quantitative questions in plain English. The full workflow requires no custom code, takes under an hour to set up using Google Sheets, and can be extended to Postgres or MySQL with a credential swap.
How to Use NotebookLM for Business Analysis
Google NotebookLM is a document-grounded AI assistant that draws answers only from the files you upload, citing the exact source page for every response. You can upload PDFs, reports, and business documents, then ask questions in plain English to extract key metrics, compare information across files, and generate structured data tables. The free tier supports up to 50 sources per notebook. NotebookLM Plus starts at $7.99 per month.
How to Get Started with Hex for Data Analysis
Hex is a collaborative analytics workspace that combines SQL, Python, and AI in a single browser-based environment. Teams can connect a data warehouse or upload a CSV file and start querying in minutes, with no local setup required. The Notebook Agent, which runs on Anthropic's Claude, writes and debugs code from plain-English prompts. As of early 2026, thousands of data teams use Hex daily as an alternative to standalone Jupyter notebooks.
How to Set Up Snowflake Notebooks in Workspaces
Snowflake Notebooks in Workspaces, generally available as of February 2026, replaces the original Legacy Notebooks product. The new version runs on Snowpark Container Services, supports native Git integration, and lets teams collaborate in shared workspace folders. It also introduces Notebook Project Objects, versioned units that Snowflake Tasks can schedule for automated runs. This guide walks through creating a workspace, launching a notebook, connecting to Snowflake data, and writing your first SQL and Python analysis cells.
How to Set Up Snowflake Cortex Analyst
Snowflake Cortex Analyst is a managed natural-language query layer built into the Snowflake Data Cloud that lets analysts and business users ask questions about structured data in plain English and receive SQL-generated answers without writing code. Setting it up requires a semantic model YAML file that describes your tables and business terminology, plus a Streamlit chat interface for end users. A standard setup for one or two tables takes two to three hours.
How to Get Started with Apache Spark 4.1
Apache Spark 4.1, released in December 2025, is the second major update in the 4.x series and resolved over 1,800 issues from more than 230 contributors. It introduces Spark Declarative Pipelines for defining data pipelines without managing execution graphs, a 1.5 MB pyspark-client for lightweight Spark Connect access, sub-second streaming via Real-Time Mode, and SQL Scripting enabled by default. You can install the Python client in under a minute with pip and start running queries against a local or remote Spark Connect server without configuring a JVM.