AI Software Engineering Bootcamp (Self-Paced)

Build web applications with Claude Code, GitHub, Next.js, Supabase, and APIs. Practice code review, debugging, design, testing, security checks, and deployment.

Plan, design, build, and launch a real website, with an AI assistant handling the code.

• Pick the right website type and plan its pages, content, and primary actions
• Map the site structure and create wireframes before any visual design begins
• Build responsive layouts on a flexible, consistent design system
• Improve accessibility, performance, imagery, and motion across the site
• Set up contact forms, SEO, analytics, and a custom domain
• Direct AI to build, test, and publish the finished site to a live URL

Understand the design decisions behind a real design system, then steer AI to build an interactive version that is yours.

• Build design systems around typography, color, spacing, contrast, and corner styles
• Translate visual decisions into reusable design tokens and a single source of truth
• Assess designs across web, mobile, social media, and print
• Prepare developer handoffs, Figma variables, and preview links ready for clients
• Use AI to plan, build, review, and refine a custom design tool

• Translate a bookkeeping workflow into a data model (accounts, transactions, statements, reports) and a working database schema
• Build bookkeeping basics in code: a chart of accounts, categorizing transactions, and reconciling against statements
• Compute financial stats, revenue growth over time, and profitability margins in JavaScript, never letting the model invent a number
• Build charts, KPI cards, anomaly flags, and an exportable report from values your code calculated
• Call the Claude API from your own server-side code to write commentary, a monthly summary, and anomaly explanations grounded in the actual figures
• Build an Ask-Your-Data copilot: load a CSV, ask in plain English, and have Claude select the chart while your code handles the math
• Keep your Claude API key server-side, handle AI failures gracefully, and persist data to a hosted database
• Deploy the finished dashboard to a live URL with production secrets and a pre-launch checklist

Build a real AI marketing studio that pulls in your analytics, has Claude read them, and generates on-brand campaigns and imagery, then deploy it live.

• Design a marketing data model and hosted database for campaigns, assets, analytics, and content
• Build a multi-screen app with Next.js, Tailwind CSS, and Claude Code
• Connect Google Analytics to surface real campaign performance data
• Use the Claude API to turn analytics into clear, actionable insights
• Generate on-brand email campaigns and images with Claude and Gemini
• Secure API keys, save content, handle errors, and deploy the finished app

Build the back office for any client-based business: clients, scheduling, AI session notes, and billing, all in one real app you deploy.

• Translate business needs into a database for clients, sessions, notes, and invoices
• Build a secure CRUD app with Next.js, Tailwind CSS, Postgres, and user login
• Track client packages, session status, attendance, hours, and remaining balances
• Use the Claude API to tidy up notes, create summaries, and suggest next-session goals
• Manage invoices, overdue balances, and editable AI-generated payment reminders
• Secure client data, deploy the app, and complete a pre-launch QA checklist

• Explain the architecture of a chatbot application (front end, server, database, LLM integration) in plain language precise enough for an AI agent to act on
• Steer an agent to scaffold and connect each of those pieces, and confirm the result actually works instead of just "looking done"
• Load a company's own documents into a chatbot's knowledge base so it answers from that material and not from general knowledge
• Separate authenticated users from general users and steer an agent to implement that distinction, including persisted chat history
• Design and test guardrail logic that keeps a chatbot on subject and turns away unrelated questions
• Specify and review agent-built feature additions, including a human escalation path with an email notification integration, and iterate on what testing shows
• Assemble a basic evaluation set to check whether a chatbot stays grounded and on subject, and use it to catch regressions
• Deploy a working version of a chatbot to a live environment

Plan and build a full-stack application with AI-powered features. Practice testing and deployment, document your work, and present the project.