Enrolments close June 1
Delegate real work to AI
Build practical AI workflows around the tools, files, and decisions that shape your week.
AI Level Up teaches you to hand off research, writing, admin, email, files, and operations work to AI tools you can actually control, review, and reuse.
Agentic AI tools can act on your behalf - but only if you know how to set them up. This course teaches you to create your own AI workflows that handle tasks the way you want them handled.
This isn't about fancy integrations or expensive tools. It's about learning to work with agentic AI tools that can act autonomously on your behalf.
So, if you want to stop copy-pasting between AI and your tools, and learn to create workflows that work the way you need them to, this course is for you.
From chat to delegation
Most people use AI as a clever chat box. They ask, copy, paste, and still do every step themselves. This course teaches you to manage AI like a practical assistant.
Before
AI gives advice
- •Ask ChatGPT or Claude for help
- •Copy the answer into another tool
- •Manually do the work yourself
- •Repeat the same prompts next week
After
AI prepares the work
- •Give AI the right context
- •Let it read, organise, draft, and prepare
- •Review before anything important happens
- •Save the routine so it improves over time
Meet your new personal assistant
Imagine starting the week with an assistant who already knows your priorities, can read the right files, prepare the next draft, organise the messy notes, and surface the actions you need to decide on.
This course helps you build that working relationship with AI. You will learn how to give it context, delegate clearly, review its work, and turn useful one-off tasks into repeatable routines.
The result is not just a few clever prompts. It is a different way to run your week: less time copying, sorting, rewriting, and chasing admin, and more time on the work that actually needs you.
Get ahead by getting the right work off your task list.
Create AI workflows in 6 weeks
What you'll be able to do:
Stop copy-pasting
have AI work directly with your email, calendar, and files
Automate the boring stuff
build workflows that handle repetitive tasks your way
Never lose a note
create an AI-powered knowledge system for research and ideas
Connect anything
link AI to Gmail, Slack, Notion, or any other software you use
Stay in control
own your data and switch providers whenever you want
Make it yours
build custom workflows tailored to how you work
I'll show you how to do it safely
When you give AI the ability to act, you're also giving it access.
You'll learn what you actually need to know when AI has real access to your tools: your email, calendar, files, and workflows.
This is why I'm such a fan of owning your own workflows rather than signing up for some company you've never heard of. Who knows how trustworthy they are? Who knows how they're storing your credentials?
A big focus of this course is how to do this stuff safely.
Ready to stop copy-pasting and start delegating?
Choose your tier6 weeks to practical AI delegation
Each week turns a real part of your work into clearer context, better delegation, safer review, and reusable workflow assets.
1Choose your AI setup and ship a first workflow+
Core path
Pick a desktop AI agent, understand provider / harness / model, and delegate one real task from your work.
Build path
Set up Claude Code, Codex CLI, GitHub, and a project workspace for more configurable workflows.
2Build your context system+
Core path
Create a practical home for the context AI needs: preferences, examples, notes, workflows, and reusable instructions.
Build path
Version-control your context hub and learn the Git habits that make deeper automation safer.
3Delegate safely+
Core path
Decide what data AI can work with, where you need a human check, and which actions should only happen after review.
Build path
Configure permissions, MCPs, CLIs, and stronger tool boundaries for higher-control workflows.
4Turn repeated work into reusable routines+
Core path
Package recurring tasks into saved instructions, checklists, templates, and repeatable project folders.
Build path
Build portable skills and agent instructions that work across CLI agents and shared project repositories.
5Use focused AI assistants+
Core path
Separate research, drafting, review, and action so each assistant has a clear job and a clear review gate.
Build path
Configure subagents, scoped tools, and deeper agent workflows for parallel or more complex work.
6Build from your automation hitlist+
Core path
Use your highest-value tasks to create practical workflows you can keep refining after the course.
Build path
Script, browser-automate, or vibe-code the workflows that deserve a custom tool or interface.
BONUS: Ditch Your Expensive SaaS
There's amazing open source software out there - CRMs, project management tools, analytics platforms - but it takes a bit of skill to set up and run.
AI changes that. With AI assistance, you can install and manage open source tools that would have required a developer before.
Take Twenty - the #1 open source CRM. Instead of paying hundreds per month for Salesforce or HubSpot, you can run your own CRM with full control over your data.
AI doesn't just save you time - it saves you money by replacing expensive software you're renting with tools you own.
Things you'll learn
The technical skills that let you get AI doing useful stuff
Agents vs workflows - what's the difference and when to use each
Design, build, test - the full lifecycle for agents and workflows
Tool calls - how AI actually takes action in the real world
Software connections - link AI to Gmail, Slack, Notion, and more
Local LLMs - run models on your machine, skip the cloud
Security - managing risk when AI can take actions
Prompt engineering - craft prompts that work reliably
Debugging - what to do when workflows don't work
Is this for you?
This is for you if...
- •You spend time each day on repetitive digital tasks
- •You want more from AI but don't know the next step
- •You're tired of copy-pasting between AI and your tools
- •You value data sovereignty and provider flexibility
Not for you if...
- •You're brand-new to ChatGPT/Claude and don't use it at least a few times a week
- •You want a library of prompts, not a system for repeatable workflows
- •You're not ready to ship real projects during the program
- •You expect automation without learning how to structure inputs and outputs
Prerequisites
- •You use an AI tool at least a few times a week
- •You can install a desktop app on the computer you will use for the course
- •A Claude, OpenAI, or Google AI subscription is recommended. We will help you choose the right setup.
- •A real workflow from your week that you want to improve
Choose your learning tier
Start delegating real work to AI with confidence and control
All prices are in USD - contact us if you want to pay by bank transfer
Sending 3 or more people? Get in touch about team pricing.
Your instructor
Joshua Vial
Founder, Code With JV
I've spent 25 years writing code and 12 years teaching people how to build software. I co-founded Dev Academy in 2013, New Zealand's leading software development bootcamp, and have been consulting and teaching independently since 2023.
These days I'm focused on one thing: helping people get real, practical value from AI. Not the chatbot stuff, but workflows and tools that actually handle work for you.
This course is designed to be as hands-on as possible. You'll leave with workflows you built or mapped on your own real work, not just notes from a slide deck.
What People Say
From participants in AI Level Up and related workshops.
“Now I feel like I have a foundation and can learn or build what I need with AI assistance going forward. The result was confidence in picking up new things just beyond my learning edge.”
Alanna Irving
Co-founder, Fundsorter
“Between daily briefings, automating contracts register updates, building research pipelines for long-form content, and tone-of-voice skills, I'm building actual things.”
Abigail Little
GM, MadeCurious
“Great course, I have learned heaps. I now understand how agentic AI can be used collaboratively in a team, and I have created and applied skills for repeating use cases.”
Andreas Benz
Senior Product Designer, Infometrics
“We all thought the course was super valuable. People got distracted from the course content because they were so busy applying what they had already learned.”
Rob Holmes
Discovery Director, PaperKite
What you'll get access to
Your own dedicated learning centre with easy-to-access modules and clear, bite-sized lessons. Key resources like PDF slide decks, written context, and next steps for every lesson plus clear visual progress as you complete each weekly module.
Your learning centre
Lesson view
You won't be doing this alone
AI Level Up includes a live community and regular Q&A sessions so you always have somewhere to turn when you get stuck or want to go deeper.
Community Discord
Join a private Discord server with your fellow students and JV. Ask questions, share what you're building, get unstuck fast, and stay accountable throughout the 6 weeks.
Weekly Live Q&A
Weekly group calls on Google Meet with JV. Bring your questions, see live demos, watch problems get solved in real time — and get answers tailored to what you're actually working on.
Wednesdays at 8:00pm NZT · Recordings available
Got Questions?
How much time do I need?
3 to 5 hours per week. The course is designed to fit around your schedule with pre-recorded content and weekly live sessions. Most of that time is hands-on, building workflows that add value to your work.
Which AI tool should I use?
If you are choosing fresh, start with Claude Code. If you already have a strong preference, OpenAI Codex Desktop and Google Gemini are supported alternatives. The course teaches portable patterns, not one fragile UI trick.
How does the build path fit in?
Some workflows are best handled by better delegation inside a desktop agent. Others deserve a custom script, browser automation, a shared repository, or a small internal tool. The build path gives you those options when they create real leverage.
When are the live calls?
The weekly group call is scheduled for Wednesdays at 8:00pm NZT / 08:00 UTC. Check your local time. Recordings will be available for everyone who cannot make it live.
When will the course launch?
The self-paced Essentials version is open now. The Plus live cohort begins with the first live call on Wednesday 3 June at 8:00pm NZT / 08:00 UTC.
Have more questions? Get in touch.
