AI-Assisted Python training course

Highlights

Live through one complete agentic build in a single day — from a vague idea to a working, tested command-line tool
Build a real log-file analyzer CLI that reports active clients, status-code breakdowns, and time-windowed request rates
Practice spec-driven development: turn fuzzy requirements into a clear specification before any code is generated
Compare multiple AI-generated proof-of-concept approaches before committing to a design
Apply test-driven development (red-green TDD) with an agent so you can trust code you didn't type yourself
Learn to keep an AI agent on target using plans, checkpoints, and review gates to prevent scope creep
Understand what genuinely changes — and what stays the same — moving from traditional to agentic development
Hands-on throughout: build alongside the trainer with shared checkpoints and optional stretch goals
No local installation needed — a ready-to-use JupyterHub environment is provided

Course Details

What you'll build

A real log-file analyzer CLI in Python
Parses web server logs to report the most active clients
Breaks down requests by status code
Calculates time-windowed request rates
Scoped to finish in a day, real enough to exercise the full development lifecycle

What you'll learn

Spec-driven development — turning a fuzzy idea into a clear specification before any code is generated
Comparing alternatives — generating and evaluating two or three AI proof-of-concepts before choosing one
Test-driven development with an agent — red-green TDD for code you trust even though you didn't type it
Keeping the agent on target — plans, checkpoints, and review gates to prevent scope creep
What genuinely changes — and what stays the same — moving from traditional to agentic development

Format

Duration: 1 day, live and trainer-led, fully online
Style: hands-on — you build alongside the trainer throughout, not a passive demo
Project-based: one real project carried from idea to working software
Inclusive pacing: shared checkpoints keep the group in sync, with optional stretch goals for fast finishers

Tools & environment

Built using Claude Code and a structured "superpowers" agentic workflow
No local installation required — a ready-to-use JupyterHub environment with everything pre-installed is provided
Delegates authenticate with their own Claude subscription at the start of the day
Prefer your own machine? A setup guide and a one-command verification script are provided in advance, plus an optional office-hours slot the day before

Who should attend

Python developers at any level who want to work professionally with AI coding agents — whether brand new to agentic tools or already using them informally. The course suits developers looking to move from ad hoc AI-assisted coding to a disciplined, repeatable workflow. Shared checkpoints and optional stretch goals keep the day valuable for both newer and more experienced participants.

Prerequisites: Working knowledge of Python (any level beyond beginner); a Claude subscription that includes Claude Code (delegates sign in with their own account).

Feedback

4.8 out of 5 average

"Our tailored course provided a well rounded introduction and also covered some intermediate level topics that we needed to know. Clive gave us some best practice ideas and tips to take away. Fast paced but the instructor never lost any of the delegates"

Brian Leek, Data Analyst, May 2022

Sign up for the JBI Training newsletter to receive technology tips directly from our instructors - Analytics, AI, ML, DevOps, Web, Backend and Security.

More about this course

AI coding agents can now write most of your code, and that shifts where a developer's value lies. Instead of typing every line, the job becomes specifying clearly, comparing alternative designs, testing rigorously, and keeping an agent focused on the right outcome. This one-day, live online workshop is built entirely around that shift, using Claude Code and a structured "superpowers" workflow to teach professional, agentic software development in practice rather than theory.

Across the day, participants build a single real project end-to-end: a log-file analyzer command-line tool that parses web server logs and reports on client activity, status codes, and request rates over time. The project is deliberately scoped to be achievable in a day while still touching every phase of professional development — ambiguous requirements, competing designs, a tested core, and safely adding new features without breaking existing functionality.

Key techniques covered include spec-driven development (clarifying requirements before generating code), evaluating multiple AI-produced proof-of-concept approaches before choosing one, applying test-driven development with an agent to build trustworthy code, and using plans, checkpoints, and review gates to prevent scope creep or feature drift. While the workshop uses Python and Claude Code as the vehicle, the underlying principles apply to agentic software development in any language or tool.

Delegates don't just watch a demo — they build alongside the trainer throughout the day, with regular checkpoints to keep the group in sync and optional stretch goals for those moving faster. No local installation is required, as a ready-to-use JupyterHub environment with everything pre-installed is provided; those who prefer their own machine can use a supplied setup guide, a one-command verification script, and an optional office-hours session the day before.

The course is led by Mike Müller of Python Academy, who brings over 25 years of Python experience, more than 20 years of professional Python training delivery, and daily hands-on practice with agentic coding tools.

FAQs

Yes. The course content is continuously reviewed and updated to reflect the latest developments in AI-assisted and agentic software development, including updates to Claude Code, agentic workflow patterns, prompt engineering techniques, and evolving best practices for test-driven development with AI agents. Delegates learn the professional workflows that reflect how developers are actively using AI coding tools in production environments today.

The AI-Assisted Python training course is a one-day, live online instructor-led workshop that teaches Python developers how to work professionally with AI coding agents. Participants build a complete, tested command-line tool from scratch using Claude Code and a structured agentic workflow — covering the full development lifecycle from ambiguous requirements through to working, tested software. The course focuses on professional techniques for specifying, steering, and testing AI-generated code, rather than on informal or ad hoc prompting.

Participants build a log-file analyzer command-line tool in Python. The tool parses web server log files and reports on the most active clients, breaks down requests by HTTP status code, and calculates time-windowed request rates. The project is deliberately scoped to be achievable within a single day while exercising every phase of professional software development — from clarifying ambiguous requirements and comparing competing designs, through to a tested, working application. Participants build alongside the trainer throughout, with shared checkpoints and optional stretch goals.

Spec-driven development is the practice of turning a vague or ambiguous requirement into a clear, structured specification before any code is generated. In an AI-assisted development context this matters because the quality of the specification directly determines the quality of the code an AI agent produces. The course teaches participants how to interrogate unclear requirements, surface assumptions, and write specifications that give AI agents enough clarity to produce trustworthy output. This is a core professional skill for working effectively with agentic coding tools.

Test-driven development (TDD) with an AI agent involves writing tests before instructing the agent to generate implementation code, then verifying that the agent's output passes those tests before accepting it. This red-green TDD workflow is particularly important when working with AI-generated code because developers cannot rely on the same intuitive understanding of code they did not write themselves. Having a passing test suite provides the evidence needed to trust AI-generated code and to safely extend or refactor it over time. The course applies TDD as a central discipline throughout the build project.

Claude Code is an agentic coding tool developed by Anthropic that operates in the terminal and can read, write, execute, and test code autonomously within a defined project context. It is used in this course as the primary AI coding agent. Delegates are required to have a Claude subscription that includes Claude Code access, as they will authenticate with their own account at the start of the day. No local installation is needed for the course environment — a ready-to-use JupyterHub environment is provided with all required tools pre-installed.

Participants need a working knowledge of Python beyond complete beginner level. The course is not an introduction to Python — it is focused on how to work professionally with AI coding agents using Python as the vehicle. Developers at any level beyond beginner are welcome, whether they are new to agentic tools or have already started using them informally and want to adopt a more disciplined, professional approach. Shared checkpoints and optional stretch goals ensure the day is valuable for both newer and more experienced participants.

This course is not an introduction to Python or a general Python skills course. It is specifically focused on the professional workflow for working with AI coding agents — how to specify requirements clearly, compare alternative AI-generated designs, apply test-driven development to code you did not write, and keep an AI agent focused and on target throughout a real project. The Python language is the vehicle for the build project, but the techniques and principles covered apply to agentic software development in any language or tooling environment.

The course is led by Mike Müller of Python Academy, who brings over 25 years of Python development experience and more than 20 years of professional Python training delivery. Mike has daily hands-on practice with agentic coding tools and has designed this course specifically around the professional workflows that experienced developers use when working with AI agents in practice.

Build a real, tested Python tool from scratch using a professional agentic workflow with Claude Code — learn to specify, compare, test, and steer AI-generated code like a professional, not just prompt it.

Public Courses

Customised Courses

Highlights

Course Details

Who should attend

Feedback

4.8 out of 5 average

Newsletter

More about this course

FAQs

Highlights

Course Details

Who should attend

Feedback

4.8 out of 5 average

Newsletter

More about this course

FAQs

Is this AI-Assisted Python course kept up to date with the latest agentic tools and workflows?

What is the AI-Assisted Python training course?

What will I actually build during this course?

What is spec-driven development and why is it taught in this course?

What is test-driven development with an AI agent and why does it matter?

What is Claude Code and do I need a subscription to attend?

What Python experience is required to attend this course?

How does this course differ from general Python training?

Who leads this course and what is their experience?

Email this page to yourself