Choose a course by the skill you want to practise and the work you will produce. A planning lesson, an editable coding lab and an algorithms exercise offer different kinds of practice.
Use the engineering roadmaps to compare software, data, AI and ML paths. Begin with one path and a small finished project rather than collecting overlapping courses.
Check each app's current course scope. LLM Academy is a compact three-lesson fine-tuning planning course; it does not run a training job. Other apps offer different lesson counts and practice environments, as listed on their own pages.
| App | Main task | Website release requirements |
|---|---|---|
| AI Engineering | 400 lessons, 140 runnable code labs and an offline tutor — from first Python to a principal-architect capstone. | macOS 14 or later; Apple silicon and Intel |
| Data Engineering | 400 lessons on SQL, pipelines, orchestration and warehousing, with runnable labs and an offline tutor. | macOS 14 or later; Apple silicon and Intel |
| Software Engineering | 400 lessons on the craft after syntax — design, testing, review, debugging — with editable labs, offline. | macOS 14 or later; Apple silicon and Intel |
| ML Engineering | 400 lessons from Python to PyTorch, serving and MLOps, with runnable labs and an offline tutor. | macOS 14 or later; Apple silicon and Intel |
| DevMonkey | Learn web development by shipping pages — graded against the real rendered DOM, not your source text. | macOS 14 or later; Apple silicon and Intel |
| CodeMonkey Pro | Trains the judgement after syntax — refactoring, review, debugging — across eight deterministic studios. | macOS 14 or later; Apple silicon and Intel |
| CodeMonkey: Python Quest | A puzzle game driven by real Python running on-device — a bug shows up as the character doing wrong. | macOS 14 or later; Apple silicon and Intel |
| Algorithms Arena | Drive algorithms step by step through interactive traces, then reinforce the patterns with drills, boss battles, code assembly and local spaced review. | macOS 14 or later; Apple silicon and Intel |
| LLM Academy | Plan the model, dataset and first efficient fine-tuning configuration through three offline lessons with local checkpoints and private notes. | macOS 13 or later; Apple silicon and Intel |
Each website app costs $10 once, or choose two different apps for $20 or every published app for $49. Optional third-party services may charge separately. Check the linked release record for the exact version and installer.
A 400-lesson course that runs as a Mac app rather than a website. Code labs are editable and run in place, the 80 architecture decisions make you choose before showing you the trade-off, and the tutor is grounded in the curriculum and works offline.
Consider before buying: It teaches engineering judgement, not one vendor's certification syllabus.
See AI Engineering screenshots and requirementsThe same 400-lesson format applied to data engineering: ingestion, modelling, orchestration and warehousing, with runnable labs and a curriculum-grounded offline tutor.
Consider before buying: Vendor-specific consoles change constantly; this teaches the transferable shape, not this quarter's UI.
See Data Engineering screenshots and requirementsA 400-lesson software engineering course as a native Mac app, covering the craft after syntax — design, testing, review, debugging — with editable labs and an offline tutor.
Consider before buying: It will not teach you a specific company's internal framework.
See Software Engineering screenshots and requirements400 lessons from first Python through PyTorch, training, serving and MLOps, as a Mac app with runnable labs and an offline curriculum-grounded tutor.
Consider before buying: Training large models needs hardware this app cannot conjure; it teaches the method, not the cluster.
See ML Engineering screenshots and requirementsTeaches web development by making you build pages. HTML, CSS and JavaScript run in a sandboxed local WKWebView with a live preview and a captured console, and the checker inspects the actual DOM and computed styles rather than string-matching your source.
Consider before buying: No backend. It teaches the front end thoroughly and stops at the network boundary.
See DevMonkey screenshots and requirementsTrains the judgement that comes after syntax — reading unfamiliar code, refactoring safely, reviewing diffs, debugging, mutation testing, system design and security review — across eight studios with deterministic grading, running Python and JavaScript locally.
Consider before buying: It assumes you can already program. This is the layer above syntax, not an introduction.
See CodeMonkey Pro screenshots and requirementsA puzzle game where the controller is real Python. Your code runs through embedded CPython on the device, produces an action stream, and the game replays that stream in a SpriteKit world — so a bug is visible as the character doing the wrong thing.
Consider before buying: It teaches programming logic through puzzles. It is not a path to a production codebase on its own.
See CodeMonkey: Python Quest screenshots and requirementsA complete offline algorithms and data-structures course for Mac with 17 chapters, interactive Watch, Predict and Perform traces, spaced review, drills, CodeBuild challenges and bounded practice labs.
Consider before buying: Algorithms Arena is an educational practice tool, not an accredited course, interview simulator, examination, employment service or guarantee of academic, hiring or professional outcomes. Its explanations, invariants, complexity claims, answer keys and drills are checked learning material but still require independent judgment, especially when applying a pattern to production code. Your Input is deliberately bounded and accepts data rather than arbitrary code: bubble sort supports at most 64 whole numbers between −999 and 999, graph breadth-first search supports at most 12 nodes and 32 comma-separated undirected edges, and palindrome checking supports at most 64 visible characters. The four-language snippets are study listings, not a general compiler or IDE. Big-O device measurements use small bounded workloads; they illustrate growth and local timing rather than provide a reproducible hardware benchmark or performance guarantee. Learning progress is local to the current Mac; readable JSON export is not a claim of cloud backup, cross-device sync or a built-in import and restore workflow. The Direct edition has private local achievements and no Game Center, online leaderboards, social classroom, instructor feedback or live interview evaluation.
See Algorithms Arena screenshots and requirementsA compact offline planning course for Mac with three guided lessons on choosing a foundation model, preparing a training dataset and configuring an efficient fine-tune.
Consider before buying: LLM Academy Direct is a compact three-lesson planning course, not a training runtime, dataset tool, model manager, MLOps platform, accredited qualification or comprehensive fine-tuning curriculum. The supplied SmolLM 1.7B, LoRA and Apple-silicon facts are illustrative prompts, not a hardware benchmark, model recommendation, licence opinion or assurance that a particular experiment is feasible. The app does not inspect licence terms, calculate memory requirements, import or scan a dataset, detect duplicates or private information, install Python or package managers, download model weights, execute optimization, monitor loss, evaluate model output, compare checkpoints, export adapters or start an inference server. Learners must implement and validate every resulting plan with suitable external tools, independent licence review, data governance and evaluation. Checkpoint completion is self-marked and does not prove competence or project readiness. Notes and completion are stored locally in ordinary macOS app preferences; they are not an encrypted project vault, portable course export, cloud backup, synchronization or collaboration system. Course completion cannot guarantee a successful model, lower costs, employment, examination or professional outcome.
See LLM Academy screenshots and requirementsBrowse all app categories or use the app finder.