Claude Senior Java Engineer Interview Preparation

The course

Claude Senior Java Engineer Interview Preparation is a self-study course: you read its lessons in a browser and work its practices in an editor, on your own machine. A read-only preview of the lessons is online: see the website link at the top of this repository.

It has 166 units in 45 modules, grouped in 10 topic areas: Java Fundamentals, Java Classes & Objects, Exception Handling, Functional Programming in Java, Java Memory Model, Multithreading and Concurrency, Modern Java Features, Principles, Design Patterns and Advanced Java Topics.

It holds 906 practices: 743 to write in code, graded by a real runner, and 163 short quizzes, graded in the page.

What you get

Lessons. Each unit is a page of text and code, with the course contents, an outline of the page and links to the previous and next unit. The index shows what you have read and what comes next, and filters the contents as you type.

Practices graded by a real runner. A code practice gives you a statement, starting code and tests. You write your answer in an editor, and Submit runs the practice's tests in a runner that has no network. The report says whether the main ask is done and how many edge cases pass, and a practice counts as passed when its tests pass.

The practice workspace. Opening a practice fills the window: the statement on the left, the editor on the right, and a report bar below. Drag the dividers to resize the panes. A worked solution is one click away when you are stuck. Run tries your code, Submit runs the tests.

Quizzes graded in the page. A short quiz has a few questions with one answer each. Choose, press Check answers, and each choice is explained right or wrong. Nothing is sent anywhere: the page grades it.

Code examples that open in an editor. A lesson's example expands in place into the editor, with its source and its test beside the text and a button that runs the test. The editor shows a copy, so the course's own files stay as they are.

Reading marks and progress. Mark a unit as read and the index counts it. Your marks and quiz passes are kept in your browser; your practice results are kept on your machine, in a Docker volume.

Optional narration. Lessons can be read aloud, the passage being read highlighted, with play and pause, previous and next passage, and speed. Practices and code are not narrated. See Narration.

Two ways to use it

The read-only preview is a copy of the lessons on a website: you read, answer the quizzes and keep your reading marks, and everything that needs a server on your machine says so and points here. The full course runs on your machine with Docker.

	Read-only preview	Full course, with Docker
Set-up	None: open the website link at the top of this repository	Docker, and one command: Run it locally with Docker
Lessons, the index and its filter	Yes	Yes
Reading marks and progress	Yes, kept in your browser	Yes, kept in your browser
Quizzes, graded in the page	Yes	Yes
Run and Submit, graded by the runner	No	Yes
The editor beside the statement	No	Yes
Code examples opened in the editor	No: the files open in the source viewer	Yes
Narration	No	Optional
Works offline	No	Yes

What you need

Docker: Docker Desktop on Windows or macOS, or Docker Engine with the compose plugin on Linux. Nothing else: no Java, no Python, no other download. What the practices need (Java and Maven) is inside the runner. A current browser. Your machine needs room for the images, which download once.

Run it locally with Docker

1. Install Docker, and start it.

2. Clone this repository and open a terminal in its directory.

3. Copy course.env to .env, and set STUDYFORGE_NAMESPACE in it to the Docker Hub account sorkhpar: the images are published under that account.

   cp course.env .env
   

   On Windows, use copy course.env .env.

4. Start the course from the published images:

   docker compose -f compose.pull.yaml up -d
   

   The first start downloads the images, which takes a while. Until STUDYFORGE_NAMESPACE is set, compose stops and says so.

5. Open http://127.0.0.1:8772/ in your browser.

The images are built for amd64 (Intel and AMD). On an Apple Silicon Mac they run under emulation, which is slower.

To build the course's own images from this checkout instead, set STUDYFORGE_NAMESPACE first: the shared base images are pulled from that account. The first build takes a while, and downloads only those bases and the course's pinned dependencies:

docker compose up -d --build

To see what is running: docker compose -f compose.pull.yaml ps. Use the same -f compose.pull.yaml with every compose command for the pulled course.

The ports

What	Where
The study site	http://127.0.0.1:8772/
The editor, which the site opens inside each practice	http://127.0.0.1:8444/

Both listen on this machine only. To use other ports, copy course.env to .env and change COURSE_SITE_PORT and COURSE_EDITOR_PORT.

Where your work is kept

Your answers, your progress and the editor's settings live in Docker volumes, so they survive a restart. down stops the course and keeps them; down -v deletes them too. Every setting is explained in course.env.

Narration (optional)

The site is complete without narration: a lesson with no recording shows no player. The recordings are not stored in this repository, because of their size: they are assets of this repository's release, and the pulled voiced site already has them. To hear the lessons read aloud, set COURSE_NARRATION=with-narration in .env (copy it from course.env), then:

• Pulled images: docker compose -f compose.pull.yaml up -d pulls the voiced site. Leave the setting as it is for the site without narration.
• Building it yourself: download the recordings from the course's release first, then build:

sh .studyforge/narration-release/restore.sh
docker compose up -d --build

On Windows, run pwsh .studyforge/narration-release/restore.ps1 instead of the first line. The scripts are .studyforge/narration-release/restore.sh and .studyforge/narration-release/restore.ps1. The script checks every download and every recording against the checksums in this repository before it places anything, and deletes the downloads afterwards. Set NARRATION_LOCAL_DIR to a directory holding the release's volumes to read them from disk instead of downloading them.

The exercises

exercises/ holds the course's authored exercises: each one's statement, starter, tests and reference solution. practice/ holds your working copy of each: the files you edit and submit live there, and exercises/ is only read.

Stop it

docker compose -f compose.pull.yaml down

(docker compose down stops a course started with the build file.)

The online preview

The read-only preview is built automatically from main on every push, so it is never out of date, and no other branch holds it. To switch it on for your copy of this repository, open Settings, then Pages, and choose "GitHub Actions" as the source. The next push to main publishes it, and the Actions tab can run it by hand. The website link at the top of this repository opens it. The workflow is .github/workflows/pages.yml.

Licence

The course's licence is in LICENSE.

How this course was built

How it was built lives on the studyforge/build branch; you do not need it.