AI in the Heartlands
What would my neighbour build, if they knew how to build with AI?
That's the question we set out to answer when making AI in the Heartlands. In our opinion, AI is a tool the way a spreadsheet or a web browser is a tool; and since we had some experience making stuff with AI, why not spread this knowledge and see where it goes?
So we're launching with our course materials free while we run workshops alongside it where you can learn this stuff together with and meet other like-minded people who want to create using AI.
Let's learn it together?
Learning new things with others can be way more productive and fun than going at it alone. Also helps that when things break, you can ask us and we can explain to you what's actually happening instead of you telling Claude/ChatGPT to "fix it please"
We are still lining up dates and venues. Follow our calendar and you will see a session the moment one opens. The course itself is ready whenever you are.
Build things your community can use
AI in the Heartlands teaches you the same skills that industry professionals use when leveraging AI, except we teach this in the context of building things for yourself and your community. For example, what if there was a HRMS-like software for your block which can help manage things like community get-togethers? What if you had an agent that reminds you your child's first performance is in two weeks time?
Installing the same tools professionals use. All MacOS, Linux, and Windows are supported; we recommend a Mac though!
Agents are the foundation of how we build things with AI. Learn how to set one up, what inputs we have to configure them, and further along, we'll learn how to work with multiple of them in parallel
Learn to use agents to create static websites and host them for free. Think personal portfolios, personal websites, or even simple tools that work in your browser
Learn just enough basics of software engineering to allow yourself to build platforms which remember data. Think creating your own blogging platform, or a newsboard for your community, or even a neighbours-only Carousell clone
Beyond building things for others, AI can also be used to improve our own lives by integrating tools and data which we use into automations that we can run for ourselves. Learn how to create a system that creates daily reports about things you care about: local news, stock markets, even social sentiments.
The (in)famous last mile problem of today's AI-developed applications. We've run global production systems in regulated environments before, and we're scaling this experience down to teach you how to run your own systems securely enough that you'd even breeze through SOC 2/ISO 27001 compliance audits
All materials are free to read and learn from, and they are public on GitHub. Teaching them, or reusing them in other course materials, needs our written permission; see the Terms of Use for more information. You can also engage us to teach these materials if you have a group 👇🏼
Already have a group who wants to learn?
We can run this course for existing groups, and we can also licence the materials to people who would rather teach it themselves.
- Residents' networks, RCs, community centres, libraries, and interest groups, we welcome you!
- We can come to your space or we can rent an event venue. A community hall or even someon's place with power points and decent wifi is enough.
- We keep group sizes small, so nobody gets left behind at the install step.
- We adjust the pace and the examples to suit the people in the room.
- A practical first step for teams who have dabbled with AI but haven't yet had clear direction or instruction.
- Run at your office or at an event venue, we do this typically in 3-4 hour blocks
- We can contextualise examples to the tools your team already subscribes to.
- Best for non-engineering teams with deep domain expertise; that is who the course was written for anyway!
/RUN_A_SESSION/
Want this course run for your residents' network, your community centre or your team? Or want to licence the materials and teach it yourself? Tell us roughly who the group is and where you are, and we will come back with options.
Prefer good ol' email?
/HELP_MENU/
Short answers to what people ask us most, before they sign up for a session.
A ChatGPT Plus subscription, around $30 a month. You pay OpenAI directly — we do not resell it. A Claude Pro subscription is optional; we find Claude better at fixing code, but you can finish the course without it.
A laptop with at least 8GB of RAM, and full administrator rights on it. Not a work laptop: the course installs third-party software, and what you run on a company machine is your employer's risk and yours.
A few hours, and a willingness to type a command and see what happens.
Each session lists what it covers and what it costs on its own page on our calendar, at luma.com/aijutsu.
We keep the rooms small on purpose (20 people maximum); this works better to ensure nobody gets left behind at the install step or is too afraid to ask questions.
No. The whole course is free to read at https://aith.aijutsu.dev. There is no sign-up and no account needed.
Workshops are for people who would rather be guided through it with others in the room. You are not buying access to the material; you are buying the room, the pace and the help.
No. The course is written for people who are not programmers, in plain language. Every command is written out in full, for both Mac and Windows.
New words are explained as they come up, and there is a glossary on the course site if one slips past us.
That depends on which workshop you're attending. We cover the full range of possibilities from a customised agent, to static websites, to full blown SaaSes, to workflows. The software you build will be scaled down for the sake of teaching it, but the same techniques you learn can be used to scale it up after the course is over.
Seven are planned in total, and one is written: course 001, building community agents with NanoClaw. The overview above marks which areas are ready now and which are still planned.
We would rather show you the whole map than let you assume the first course is the whole thing. The planned ones are plans, not promises, and the course site always shows exactly what exists today.
The materials are free to read and learn from, and they are public on GitHub. Teaching them, or reusing them in other course materials, needs our written permission.
Licensing is a short conversation. Email [email protected], or use the form above.
Yes. We run it for residents' networks, RCs, community centres, libraries and interest groups, and for companies who want a practical first step into AI agents.
We come to your space — a community hall with power points and decent wifi is enough — or we run it at our office. Tell us roughly who the group is and where you are using the form above, and we will come back with options.