AI guides and hands-on labs
-
Run a broker and its specialist agents from one config file
One XML file defines an agent network's specialists, routing prompts and risk rule. Run it in Colab, watch the human gate fire, then swap the file.
Source: Agent network
CTOAI EngineerDeveloperIntermediateAI control plane
-
Catch a model that is 88.7% accurate and misses almost every case
A survey model was 88.7% accurate and found almost no one. Learn the checks that expose a model like it, then run them on card fraud data in Colab.
Source: Visual Handbook
AnalystDeveloperData ScientistBeginnerMachine learning
-
Quote a survey figure with its weights and margin of error
One survey gives 11.29% or 8.44% for the same measure, depending on its weights. Learn to weight a survey and state its margin of error, in Colab.
Source: Visual Handbook
AnalystCIOCTOBeginnerStatistics
-
Choose the right statistical test in four questions
Four questions about your data pick the right test: outcome, groups, design and assumptions. Learn the map, then run it on clinic data in Colab.
Source: Visual Handbook
AnalystBeginnerStatistics
Where the labs live
Visual guides and Colab notebooks
Python, Visually
A library of Python guides, each a set of numbered notebooks that open in Colab with nothing to install. The guides from the language basics through files, APIs, testing and databases are complete, and more library guides are planned.
Code you can run
A config-driven agent network
A broker sends each question to specialist agents and holds risky answers at a human approval gate. The roster, the prompts and every agent’s reply live in one XML file, so you change the network by editing that file.
A live run needs your own OpenAI API key and costs a fraction of a cent a question. Its 19 tests run against a stub model, with no key and no cost.