How-To Hub · sourced & dated

How-To

Practical step-by-step tutorials with screenshots

46 stories published 100% sourced & dated

The lead

Latest first

Browse How-To by topic

Latest in How-To

How-To How to isolate AI agents in GitHub Actions with Docker

How to isolate AI agents in GitHub Actions with Docker

Run coding agents inside Docker Sandboxes on GitHub Actions: install gh-aw, configure the sbx runtime, and let an agent open a safe…

Jinultimate · Aug 28, 2026
How-To How to build a fraud model in SageMaker Canvas from Snowflake

How to build a fraud model in SageMaker Canvas from Snowflake

Amazon's step-by-step guide connects Snowflake to SageMaker Canvas, letting analysts prep data and train an XGBoost fraud model with no code.

Jinultimate · Aug 26, 2026
How-To How to build a Bedrock document classifier with agents

How to build a Bedrock document classifier with agents

A three-agent Amazon Bedrock system pairs Claude Haiku 4.5 with Titan multimodal embeddings to classify insurance documents by text and visual layout.

Jinultimate · Aug 18, 2026
How-To How to evaluate NVIDIA’s AI factory land-power-shell guarantee

How to evaluate NVIDIA’s AI factory land-power-shell guarantee

NVIDIA is backing 20-year land, power and shell for OpenAI's Ohio AI factory. Learn how to read the guarantee, its limits, and…

Jinultimate · Aug 18, 2026
How-To How to let an OpenClaw agent pay for APIs with stablecoins

How to let an OpenClaw agent pay for APIs with stablecoins

Use Amazon Bedrock AgentCore payments and the aws-agents-pay plugin to give a local OpenClaw assistant a bounded USDC wallet for x402 micro-payments.

Jinultimate · Aug 17, 2026
How-To How to stream AI agent telemetry to AgentCore from anywhere

How to stream AI agent telemetry to AgentCore from anywhere

Send AI agent traces, spans, and token metrics from on-premises or another cloud into the AgentCore Observability dashboard using the AWS Distro…

Jinultimate · Aug 14, 2026
How-To How to run automated behavioral research on AI agents

How to run automated behavioral research on AI agents

AEROBAT tested 79 hypotheses across 1,240 experiments and 23,512 simulations to map AI agent behavior—here is how its automated pipeline works.

Jinultimate · Aug 14, 2026
How-To How to build a Bedrock AgentCore M&A due diligence agent

How to build a Bedrock AgentCore M&A due diligence agent

Deploy AWS's open M&A due diligence sample on Bedrock AgentCore: a supervisor and four specialist agents screen targets, run DCF, and cite…

Jinultimate · Aug 13, 2026
How-To How to replace brittle RPA with a Bedrock browser AI agent

How to replace brittle RPA with a Bedrock browser AI agent

Build an AI digital worker that drives legacy web apps through a managed cloud browser with Bedrock and Strands Agents, with audit…

Jinultimate · Aug 13, 2026
How-To How to attribute Amazon Bedrock cost to each IAM principal

How to attribute Amazon Bedrock cost to each IAM principal

Set up CUR 2.0 with IAM principal data, query it in Amazon Athena, and read Bedrock spend per user, role, or project…

Jinultimate · Aug 12, 2026
The zBrandco Edition

Never miss what matters in How-To.