From Crypto media and technology education
Michael Gu
@BoxminingAI
AI products, agents, and emerging technology
Read Michael's storyPractical AI tutorials
Use AI agents to automate your life and build useful projects.
Follow practical tutorials for setting up agents, automating recurring work, comparing models, and turning ideas into working systems.
- 57
- tutorials
- ~11h
- total time
- 5
- topics
The instructors
Meet the team
Our instructors came from crypto media, venture growth, editorial work, and fitness. They teach the workflows that held up when AI met the jobs they already knew.
Meet the full team
From Venture building and growth
Elvin Cheung
@ekcheungAI
AI growth and agent workflows
Read Elvin's story
From SEO writing and technical explainers
Ronal Thapa
Host of AlphaDrop
Research and editorial workflows
Read Ronal's story
From Fitness coaching and content
J Martin
@martintrainhard
AI tools, automation, and independent work
Read Martin's story
Tutorial library
Browse tutorials.
Step-by-step guides grouped by the work they help you do. Open any tutorial directly or browse a complete topic collection.
View all 57 tutorials
Topic 01
Grok Bot
Build a supervised AI coworker system one safe workflow at a time.
A bounded first agent with clear instructions, permissions, and a task you can inspect.
Topic 02
Hermes Agent
Start here if you want an AI assistant that can actually do work.
A working Hermes install with a model, API key, and doctor check in the first two tutorials.
Topic 03
AI Models
Pick the right model for the job without drowning in benchmarks.
A four-tier map for matching a task to a capable, sensible-cost model.
Topic 04
Case studies
Move from watching tools to shipping something useful.
A clear choice of coding harness and the write-test-fix loop for a real build.
Topic 05
Decodo Web Scraping
Go from one safe MCP scrape to a reviewed, repeatable data workflow.
One bounded MCP scrape with source, timestamp, request count, and a result you can verify.
Model performance
Compare how AI models perform.
Give models the same brief, open their real outputs side by side, and judge quality, reliability, and how much cleanup each result needs.
Open gallery →