6.S093 – How to Ship Almost Anything with AI (IAP 2026)

MIT IAP 2026 · Jan 20–27

Units: 6 · Level: Undergraduate · Grading: P/F

This repository contains my work for 6.S093: How to Ship Almost Anything with AI v2, an MIT IAP course focused on building end-to-end autonomous AI systems.

Course Overview

The goal of the class is to learn how to design, build, and deploy agentic AI systems that can operate autonomously. The central project is a fully autonomous AI agent that manages social media for a company or individual using a “second brain” knowledge base.

The course emphasizes rapid prototyping, real deployments, and hands-on experience with modern AI tooling.

What I’m Building

By the end of the course, this project aims to include:

• An autonomous AI agent that:

  • Monitors a knowledge base (Notion / Obsidian / text files)

  • Generates and posts social media content

  • Creates custom images using diffusion models

  • Scrapes and engages with relevant online content

• (Optional features)

  • Human-in-the-loop approval via Telegram

  • Web dashboard for monitoring and control

Topics Covered

• Agentic system design & tool calling

• Multimodal LLMs and image generation

• RAG and semantic search for memory

• Web scraping and social media APIs

• Cloud deployment

• AI-powered developer workflows

Tech Stack (varies by implementation)

• Python

• LLM APIs (OpenAI / Anthropic / OpenRouter, etc.)

• Pydantic for structured tool calling

• Playwright for scraping

• SQLite / vector databases for memory

• Cloud deployment (GCP)

• (Optional) Next.js / Tailwind for frontend

Motivation

Inspired by the idea of the “one-person unicorn” — using agentic AI to build and ship full-stack products as a solo developer.

This repo is both a learning log and a working system.

Generated by ChatGPT