Setup & Tooling

For ownersThe engineer's workbench. Skim it so you know what a tidy, repeatable setup looks like when you hire one.

Get your environment ready for everything that follows.

13 lessons

Start with lesson 1

Lesson 1 of 13, ~45 minutesDev Environment

The lessons

  1. 01Dev EnvironmentSetting up a computer so it can build and run AI code: a language runtime, a package manager, an editor and a few command line tools, all installed the same way every time.Build · ~45 minutes
  2. 02Git & CollaborationGit keeps a full history of every change to code and files, so you can go back, compare versions and work with others on the same project without overwriting each other.Learn · ~30 minutes
  3. 03GPU Setup & CloudA GPU runs thousands of small math operations at once, which makes training and running models far faster than a normal chip. This lesson covers checking for one locally and renting one in the cloud.Build · ~45 minutes
  4. 04APIs & KeysAn AI API lets your code send text or files to a hosted model and get an answer back. An API key is the password that identifies your account and gets billed for each call.Build · ~30 minutes
  5. 05Jupyter NotebooksA Jupyter notebook mixes code, results, charts and notes in one page, so you can try an idea step by step and see the output right away.Build · ~30 minutes
  6. 06Python EnvironmentsA virtual environment is a separate folder of Python packages for one project, so two projects can use different versions of the same library without breaking each other.Build · ~30 minutes
  7. 07Docker for AIDocker packs an app with its exact operating system pieces, libraries and settings into a container, so it runs the same on a laptop, a server or a cloud GPU.Build · ~60 minutes
  8. 08Editor SetupConfiguring a code editor with the right extensions, formatter, linter and AI assistant so writing, running and fixing code takes fewer steps.Build · ~20 minutes
  9. 09Data ManagementHow to download, store, stream and split datasets and model files so they are easy to find, fit on disk and stay the same between runs.Build · ~45 minutes
  10. 10Terminal & ShellUsing the command line to move around files, run programs, chain commands and automate repeat tasks, which is how most AI tools and servers are run.Learn · ~35 minutes
  11. 11Linux for AIThe Linux basics needed to work on the servers where most AI runs: users and permissions, installing packages, watching memory and GPU use, and keeping jobs running after you log out.Learn · ~30 minutes
  12. 12Debugging and ProfilingFinding why code is wrong or slow: print checks, the Python debugger, logging, timing, memory profiling and spotting AI bugs that do not crash but quietly train on bad data.Build · ~60 minutes
  13. 13Python for AI EngineeringThe small set of Python habits most AI code relies on: comprehensions, functions, classes, generators, context managers, type hints, file handling, and why loops over numbers are slow.Build · ~120 minutes

Lessons from AI Engineering from Scratch by Rohit Ghumare and contributors, MIT license. Notes for owners by Lozo. License

Next step

Rather have it built for you?

Reading is optional. Tell me what's stuck and I'll build it into the business you already run.

Tell me what's stuck
  • A real conversation, not a pitch
  • You keep your accounts
  • Nothing sends without your yes