- Week 1 is about foundations and first builds: prompting, AI tools, databases, and your first deployed tool. The goal is removing the psychological barrier between 'understanding AI' and 'using it on real work.'
- Week 2 is a build sprint: you're applying everything from Week 1 to a capstone project built around your actual business or role. You ship something real.
- By the end, students leave with three deployed tools, a new professional network, and — most importantly — the judgment to evaluate and apply AI independently.
- What you won't learn: deep machine learning theory, model training, or how to build foundation models. MakerSquare is for people who want to use and deploy AI, not research it.
Most AI bootcamp descriptions are either a vague list of topics or a collection of testimonials. Neither tells you what you'll actually do on Day 3. This post is different: a week-by-week account of what MakerSquare students learn, what they build, and what changes about how they work by the end. The goal is to give you enough specificity to decide whether this is the right format for what you're trying to accomplish.
What to expect before you arrive
MakerSquare is designed for professionals with no coding background. The prerequisite is not technical knowledge — it's a real problem you want to solve. Students who get the most from the program come in knowing what they're building toward: a tool that automates a workflow in their business, a research process that takes three hours and should take fifteen minutes, an interface that doesn't yet exist for their specific industry.
Before Day 1, students get a brief pre-work assignment: set up Cursor, create a Claude Pro account, and document one workflow they currently do manually that they want to automate. That's it. The technical setup is handled on Day 1.
Week 1: foundations, first builds, and removing the mental barrier
Days 1–2: Prompting and AI fundamentals. Not theory — practice. Students spend Day 1 prompting Claude on real tasks from their own work: editing documents they've actually written, analyzing data they've actually collected, generating first drafts of emails they've actually sent. The goal is developing enough fluency with prompting that the tool stops feeling like a guessing game. By end of Day 2, every student has used Claude to meaningfully accelerate something they actually do.
Days 3–4: Cursor and first code. Students open Cursor for the first time and build something simple: a script that processes a file, an automation that connects two services, a basic interface. The point is not the output — it's the experience of directing AI to build something you couldn't have built before. Most students have a moment around Day 4 where something clicks: they realize the tool is doing the work they thought required years of programming to do.
Day 5: Databases and memory. Tools that don't remember anything aren't very useful. Day 5 introduces students to simple databases — how to store information, retrieve it, and connect it to the tools they've been building. By the end of Week 1, students have a basic deployed tool that accepts input, processes it using AI, and returns output.
Week 2: the build sprint
Days 6–8: Capstone build begins. Week 2 is a sprint. Students apply everything from Week 1 to a project built specifically around their own business or role. The instructor provides structure and unblocks problems; the students drive the direction. This is where the learning compounds: the constraints of a real project reveal what you don't yet know and force you to figure it out.
Days 9–10: Polish and deploy. The last two days are about making the project production-ready. Error handling, edge cases, interface polish, and documentation. Students deploy their project to a live environment — not a prototype, but something they can hand to a colleague and say 'this works.'
Day 10 — Demo Day: Every student presents their project to the cohort and to MakerSquare's community of alumni and Austin's tech community. The demo requirement is a meaningful forcing function: knowing you'll present publicly changes how seriously you approach the build.
What you leave with
Three deployed tools. The Day 5 project, the capstone, and usually one more thing built during the sprint that solves a specific pain point that came up during the week.
A professional network of people who built alongside you. The cohort effect is real — the professionals who went through two weeks of problem-solving together maintain those relationships after the program ends. Several MakerSquare cohort members have gone on to work together.
The judgment to keep building independently. The most important thing students leave with isn't a specific tool — it's enough experience with the AI development process to evaluate new tools, scope new projects, and debug problems on their own. That's the transferable skill.
What you won't learn
MakerSquare is explicitly not a machine learning program. Students don't train models, don't learn statistical theory, and don't study how foundation models work internally. The program is for people who want to use and deploy AI effectively in their work — not research it or build it from scratch. If the goal is to become a machine learning engineer or AI researcher, this is the wrong program. If the goal is to become the person in your organization who knows how to build with AI, this is exactly the right one.
The full MakerSquare curriculum is available to review — it covers what's taught on each day, how the capstone project is scoped, and what the Demo Day format looks like. If you're deciding whether this is the right program for your goals, that's the right place to start.