🛹 The Onramp, in a Nutshell

<aside> 🎯

This course is designed to help you get set up like a professional AI Engineer so that you’re ready to build production LLM applications in 2026.

We built it for people committed to becoming AI Engineers who don't code daily yet but want to. If that’s you, you need AI-assisted software engineering reps before tackling any intensive AI Engineering bootcamp.

This course is for you if: 1) you love building, 2) you learn best with accountability, and 3) you need access to 1:1 support with expert staff.

If you’re also a community member ready to bring the vibes into your journey group, that’s just icing on the AI Makerspace cake 🎂.

🤖 What is AI Engineering?

🙋 Why we built this course

🧑‍💻 The ideal student in pursuit of “AI Engineering”

🏅 What success looks like

💡 Real Student Example

🤔  Prerequisites

🏆 Grading

</aside>

📅 Detailed Course Schedule & Curriculum

2. Set up API Key for OpenAI GPT models
3. Set up Google Colab account

Don’t forget to using the #ask-aim channel if you get stuck!* |

Live Session 1: 🎧 From Vibes to AI-Assisted Development**

🎯 Understand how to hit LLM APIs locally/remotely, and how to use an AI IDE to modify notebooks locally/remotely according to software engineering best-practices

****- Understand course structure

• Meet your cohort, peer supporters, and journey group!
• Understand the OpenAI Chat Completions API, including it’s history and implementation
• Learn the component parts of your IDE setup, including Version Control, Command Line Interface, Package/Env Management, Python Notebooks, and Code Editor
• LLM prototyping best practices | 1. Communicate with OpenAI GPT models through the API via Google Colab Notebooks

2. Set up Cursor, our AI-assisted Interactive Development Environment (IDE), to be able to work with Jupyter Notebooks

3. Run notebook locally in IDE according to Git best-practices, using AI-assist to help understand each chunk of Python code

4. After modifications, version control notebook on GitHub and re-upload modified notebook to Google Colab as a shareable | Interactive AI App Dev Env Setup Version Control: GitHub CLI: Shell for Unix-like OS (WSL) Package & Env Management: uv Python Notebooks: Jupyter / Colab Code Editor: Cursor CLI Coding Agent: Cursor CLI

LLM App Stack Tooling LLM: OpenAI GPT models

Relevant papers AI-Assisted Development - Cursor 2.0 (Oct 2025) - Cursor Agent CLI (Aug 2025)

• Claude Code: Best practices for agentic coding (Apr 2025)
• Not all AI-Assisted Programming is Vibe-Coding (Mar 2025)

Prompting LLMs

• Chain-of-Thought (Jan 2022)
• LMs are few-shot learners (May 2020) | | **Monday, December 8, 12:00 - 1:30 PM ET

Live Session 2: ⚡ Backend Web App Development & Deployment of LLM Applications**

🎯 Understand best-practices for vibe-coding web application back ends, how to hook them up to vibe-coded front ends, and how to ensure that you can continue to work on the code!

****- What you should know about back end web app development as an AI Engineer, and when you should hand off your apps to the back end engineering /platform & infrastructure/ DevOps/MLOps team - One-shot vibe-coding of FastAPI back ends that work with vibe-coded front ends and LLM APIs

• Understand the actual level of detail necessary to set up cursor to develop a back end that you plan to modify
• Learn how to connect front ends to back ends
• Understand the role of docker containers in end-to-end deployments with FastAPI and Render | 1. Vibe-coding a FastAPI web application back end using with Cursor and deploying locally according to Git best-practices

2. Connecting our AI-assisted front end to our FastAPI back end

3. Making changes and updating our web application

4. Deploying FastAPI web apps publicly on Render | Interactive AI App Dev Env Setup Version Control: GitHub CLI: Shell for Unix-like OS (WSL) Package & Env Management: uv Python Notebooks: Jupyter / Colab Code Editor: Cursor CLI Coding Agent: Cursor CLI

****LLM App Stack Tooling LLM: OpenAI GPT models User Interface: Vibe-coded through Cursor Web App Framework: FastAPI Containerization: Docker Deployment: Render

Relevant papers/blogs

• FastAPI in Containers - Deploy a FastAPI App (Render) | | **Monday, December 15, 12:00 - 1:30 PM ET

Live Session 2: 🖼️ Connecting Frontend UIs to Backend Deployments**

🎯 Understand how vibe-coding front end user interfaces is an iterative process - and how each iteration should inform your ability to do one-shot AI-assisted development

****- What you should know about front end dev as an AI Engineer, and when to hand off to your front end/ web dev / UI/UX /product & design team - One-shot vibe-coding of front ends connected to our LLM APIs with simple prompts and making subsequent modifications directly with coding agents

• Understand the actual level of detail necessary to set up cursor to develop a front end that you plan to modify
• Learn exactly how to revert back to previous versions of the applications you build with AI assistance, minimizing AI slop
• Learn how to deploy your front end applications locally, easily, with mostly vibes, and understand how much detail we’ve yet to learn (foreshadowing…) | 1. Vibe-coding a front end using with Cursor and deploying locally according to Git best-practices

2. Reverting back to previous versions of our deployed front ends; re-deploying them after too much vibe-coding

3. Deploying our vibe-coded front end remotely with Render

| Interactive AI App Dev Env Setup Version Control: GitHub CLI: Shell for Unix-like OS (WSL) Package & Env Management: uv Python Notebooks: Jupyter / Colab Code Editor: Cursor CLI Coding Agent: Cursor CLI

****LLM App Stack Tooling LLM: OpenAI GPT models User Interface: Vibe-coded through Cursor Deployment: Render

**Relevant papers/blogs

• **What you need to know about Vibe Coding / State of Vibe Coding Report (Sep 2025) | | **Monday, December 22, 12:00 - 1:30 PM ET

Live Session 4: ↔️ Building Complex End-to-End LLM Applications**

🎯 Understand best-practices for AI-assisted development of end-to-end applications, including building with the intention of modifying and continuing to build!

****- LLM prototyping best practices, how to spin up end-to-end prototypes and vibe check them!

• Ideate with peers & peer supporters
• Build an end-to-end application solving a real problem you’re interested in using everything we’ve learned in the course! | 1. Complete The AI Engineer Challenge from scratch (without the template!) according to AI-assisted development best-practices, and using your personal Cursor and AI IDE setup

2. Define a specific problem you’d like to solve using a basic end-to-end LLM application

3. Make updates to the UI and to the developer prompt to align your application with your problem and solution

4. Redeploy your application to a public endpoint

5. Evaluate your end-to-end app through a Vibe Check!

🚧 (Optional) Advanced Build 1: Add Deep Research as an optional feature, including both the front end UI and the back end API!

🚧 (Optional) Advanced Build 2: Swap out the OpenAI API for an open-source model: for local deployments use ollama, for remote deployments use together ai!

Congratulations - you can do AI-assisted development like the pros! | Interactive AI App Dev Env Setup Version Control: GitHub CLI: Shell for Unix-like OS (WSL) Package & Env Management: uv Python Notebooks: Jupyter / Colab Code Editor: Cursor CLI Coding Agent: Cursor CLI

****LLM App Stack Tooling LLM: OpenAI GPT models User Interface: Vibe-coded through Cursor Web App Framework: FastAPI Deployment: Render

Relevant papers/blogs - The AI Engineer Challenge

• Introduction to deep research in the OpenAI API (June 2025) |