MA
MadeByAris Bootcamp
Back to home

8-week curriculum

Think clearly. Prompt clearly. Build carefully. Verify honestly.

From learning how AI actually works in Week 1, to a production capstone and demo day in Week 8. Every week ends with one outcome you can deploy or show; no empty weeks.

Duration
8 weeks
Live
48 hours

Tuesday & Saturday WIB

Cohort
Max 8
Output
4 projects

Plus your capstone

Week 1 — the most important foundation

Before you build apps, understand how the AI actually works.

The point of the first week is not "use AI to write code faster." It is to build the habit: think clearly, prompt sharply, verify honestly.

  • What an LLM actually is

    Tokens, context windows, and why the same prompt can give different answers tomorrow.

  • Prompts, rules, and skills

    How system prompts, custom prompts, project rules, and skills change what the model does.

  • Giving project context

    Files, selections, snippets, CLI paths, and how to feed only what the AI needs.

  • Plan before code

    Asking AI to plan large edits, then approving the plan slice by slice.

  • Verify honestly

    Browser, terminal, tests, and deployments. Never trust an AI claim without evidence.

  • Pricing without surprises

    How tokens, credits, subscriptions, and rate limits work across tools, so usage stays under control.

Weekly roadmap

8 weeks, 8 clear outcomes.

  1. Week 1

    AI Tools, Prompting, and Web Fundamentals

    Deployed company profile website

    The most important week. Students learn what an LLM is, how tokens, prompts, rules, and context windows shape output, and how to verify AI work in the browser, terminal, and deployment. Tool-flexible: Cursor, Claude Code, Windsurf, Cline, Aider, OpenCode, anything.

    What you build

    A one-page company profile site (header, hero, services, about, contact, footer), responsive, deployed to a live URL.

    What you can do after

    • Explain HTML, CSS, and JavaScript in simple terms
    • Use any AI IDE or CLI to plan and build a small webpage
    • Explain LLM, token, context window, system prompt, and rules
    • Write prompts with goal, context, constraints, and verification
    • Compare AI tool pricing and avoid surprise usage costs
    • Test in a browser and on mobile, commit to GitHub, deploy live

    AI workflow this week: Goal → context → constraints → verification → commit

  2. Week 2

    Next.js and Production Deployment

    Next.js site on a custom domain

    Migrate the static Week 1 site into a real Next.js project. Learn why frameworks exist, plan a migration with AI, and deploy professionally with Vercel and a custom domain.

    What you build

    A production-ready Next.js company website with App Router, reusable components, basic SEO metadata, Vercel deployment, and a custom domain on HTTPS.

    What you can do after

    • Explain the difference between static, React, and Next.js
    • Plan a framework migration with AI before editing files
    • Split a page into components without losing the design
    • Deploy to Vercel from GitHub and connect a custom domain

    AI workflow this week: Map old site → plan structure → migrate one section → verify → commit → repeat

  3. Week 3

    Tech Stack Selection and MVP

    MVP dashboard with database integration

    Move from "I can build a page" to "I can choose a stack and ship a small app." Compare tools using project constraints, write a small PRD, design a database schema, and ship a focused MVP dashboard.

    What you build

    A small MVP dashboard with a clear user, chosen stack with reasoning, database schema, basic CRUD workflow, and a README explaining the decisions.

    What you can do after

    • Compare stack choices using project constraints, not hype
    • Write a beginner-friendly PRD before building
    • Design a simple database schema
    • Build one useful dashboard workflow end to end

    AI workflow this week: Given my users, timeline, skill level, and data needs, recommend the simplest stack that can ship.

  4. Week 4

    Docker, Automation, and VPS

    Containerized app on a VPS

    Turn "it works on my laptop" into "it runs the same anywhere." Learn Docker, Compose, Makefile automation, and deploy a small service to a real Linux VPS with clear troubleshooting habits.

    What you build

    A containerized MVP with Dockerfile, docker-compose.yml, .env.example, Makefile, local container run, and a documented VPS deployment.

    What you can do after

    • Explain container vs local runtime
    • Write a basic Dockerfile and Compose setup
    • Use Makefile commands to remove repetitive typing
    • Deploy to a Linux VPS at a beginner level and read logs

    AI workflow this week: Always send AI the command, expected result, actual error, Dockerfile, and logs.

  5. Week 5

    Backend API Development with NestJS

    Authenticated REST API

    Design and build a professional backend one route at a time. Modules, controllers, services, DTOs, validation, Prisma, JWT auth, and basic API docs, with AI used to plan contracts before code.

    What you build

    A NestJS backend with modular structure, Prisma model, CRUD endpoints, DTO validation, JWT auth, a protected route, and Swagger or route docs.

    What you can do after

    • Explain modules, controllers, services, DTOs, and guards
    • Design API contracts before generating code
    • Validate inputs and return clean error responses
    • Implement basic JWT auth and test with real requests

    AI workflow this week: Model → DTO → service method → controller route → test request → commit

  6. Week 6

    Frontend SPA with Vite and TanStack

    SPA connected to the backend API

    Connect Week 5’s API to a modern dashboard SPA with Vite, React, TanStack Query, and TanStack Router. Real loading states, real error states, real auth, no fake data hiding bugs.

    What you build

    A dashboard SPA with TanStack Query, TanStack Router, login/token handling, a protected route, real loading/error states, a form wired to the API, and a deployed frontend.

    What you can do after

    • Explain client state vs server state
    • Use TanStack Query to fetch and mutate backend data
    • Build protected routes and handle auth tokens
    • Debug frontend-backend integration with real evidence

    AI workflow this week: Send AI the endpoint, method, request shape, response shape, auth requirement, loading state, and error state.

  7. Week 7

    Final Project Development

    Full capstone application

    Capstone build week. Combine the bootcamp stack into one focused, portfolio-worthy app. Use AI to plan PRD, architecture, data model, and contracts, then build feature-by-feature.

    What you build

    A capstone app with frontend SPA, backend API, database, auth, at least 2 external integrations, an SMTP feature, production deployment, and a clear README.

    What you can do after

    • Scope a capstone project realistically
    • Use AI for PRD, architecture, and task breakdown
    • Build feature-by-feature, not all-at-once
    • Debug integration issues with logs, requests, and exact errors

    AI workflow this week: PRD → architecture → data model → API contract → one feature → verify → commit

  8. Week 8

    Portfolio, Demo Day, and Career Prep

    Polished portfolio and final presentation

    Turn the capstone into a presentable portfolio story: polished README, public portfolio page, a 5-minute demo, a short developer story, and a clear next-30-day plan.

    What you build

    A polished capstone README, portfolio page, 5-minute demo script, live demo, personal developer story, and a 30-day improvement plan.

    What you can do after

    • Tell a clear story about what you built and why
    • Defend your technical choices in plain language
    • Run a tight 5-minute live demo
    • Leave with a concrete plan for the next 30 days

    AI workflow this week: Use AI for portfolio copy, demo rehearsal, and resume bullets, then rewrite in your own voice.

Tooling policy

Tool-flexible. Workflow-strict.

Cursor is the default classroom demo because it is easy to show live, but you can use any AI IDE or CLI you prefer: Cursor, Windsurf, Trae, Claude Code, Cline, OpenCode, Aider, or any other tool. The workflow is what matters: clear goals, clear context, planned edits, real verification, and committed checkpoints.

    What you must be able to do

    Give the AI a clear goal, context, constraints, and verification steps

    Use files, selected code, snippets, or CLI paths as context

    Ask for a plan before large edits

    Review the code instead of accepting everything

    Test the result with real evidence

    Commit after working checkpoints

    Understand pricing, usage limits, credits, and subscriptions

Portfolio outcomes

What you will ship to production.

Week 1–2

Company Profile + Next.js

A static one-pager and then a real Next.js site on a custom domain over HTTPS.

Week 3–4

MVP Dashboard, Containerized on a VPS

A small dashboard with a database, then containerized with Docker and Compose and deployed to a real Linux VPS.

Week 5–6

NestJS API + Vite SPA

A real backend with auth, validation, and Prisma, connected to a Vite + TanStack frontend with real loading and error states.

Week 7–8

Capstone + Portfolio Demo

A capstone app you scoped, planned, and shipped. Plus a polished portfolio page and a 5-minute demo.

Weekly schedule

Two live sessions, one consistent rhythm.

Each week has two live sessions in the same format. Tuesday is concept and demo, Saturday is build and review.

Tuesday · concept

19:00–22:00 WIB
  • · This week's concept and mental model
  • · Live demo and prompting walkthrough
  • · Guided practice and Q&A

Saturday · workshop

16:00–19:00 WIB
  • · Hands-on build, individual code review
  • · Deploy and verify this week's outcome
  • · Reflection on the most common mistakes

Ready to start with Week 1?

Seats are capped at 8 per batch so each participant still gets real code review. Apply now or talk to the mentor first.