python_course

tutor is an offline, terminal-based Python course that teaches by having you write real code instead of watching videos. The lessons, drills, worked examples, a "VIM dojo" and an animated "watch it run" tracer are all pre-authored and generated locally; the README states there is no model call and no network access, so it works fully offline and does not phone home. The course runs inside a Neovim-style modal editor. The workflow is: read the challenge, press i to start typing, press Esc, then run and submit with :!python3 % or Ctrl+Enter. A pass advances to the next challenge; a fail lets you fix the code and run again. NORMAL-mode keys include h/j/k/l movement, i/a/A/I/o/O insert, x delete char, dd delete line, yy yank, p paste, u undo, 0/$ edges, gg/G top/bottom, with Enter to dive into a lesson and shortcuts for watch, listen, lessons, quick check, cheat sheet, review, examples, volume, voice and music. Narration is spoken by a local TTS engine that is auto-detected: espeak-ng (described as the easiest and most widely available, robotic) or Piper (a neural voice installed through a provided script). F6 toggles the voice, and everything still works silently if no voice is installed. Keypress sounds and win/fail stings rely on aplay/paplay over ALSA or PulseAudio, and the installer bundles a media pack with 32 mechanical-keypress samples and a sting library. F4 opens a volume panel with separate faders for voice and sound effects, and both levels are remembered between sessions. Beyond language basics, the material includes a BUILD STUFF track with a real bash terminal and a live folder tree, so you create directories and files and then run your own program. A CLOUD & DEVOPS track is described as 11 modules and 92 lessons, covering version control with git, packaging with Docker, AWS S3, EC2, Lambda, DynamoDB and IAM through both the CLI and real boto3 Python, infrastructure-as-code with Terraform, server automation with Ansible including idempotency and handlers, and a CI/CD pipeline that goes red/green — all inside a simulated, offline cloud. Progress such as XP, level, streak and volume is saved to ~/.learning/progress.json and is untouched by updates. Installation is a one-line curl script that downloads the app to ~/.local/share/tutor/, creates a private virtualenv, installs two Python dependencies (textual and rich) and places a tutor launcher in ~/.local/bin; pip or pipx install from the Git URL are offered as alternatives. Re-running the installer is idempotent and pulls the latest code plus media pack. The project is licensed under MIT.