I built EverythingAI because I wanted to understand AI from a non-coder's point of view without avoiding the real jargon behind it. I do not learn well through traditional lessons, so I structured it more like a first day on the job with a patient senior employee beside you—someone who explains the same idea in another way until it finally makes sense.
The idea
The learner starts as the new person on the job. The path introduces the language, explains why it matters, gives examples and connects each new idea to something already understood. It does not assume that technical words are obvious.
How it works
The platform combines lessons with assessments, revision, achievements, notes, bookmarks, a glossary and search. Progress is kept locally on the learner’s computer.
Who it is for
This is for people who are curious about AI but do not come from coding, mathematics or a traditional technical background. The jargon is still taught, because understanding it can open doors, but it is explained instead of used as a gate.
Current stage
EverythingAI works locally, but I want to review the lessons and make sure the explanations genuinely help before opening it. I hope to get the full learning path running for anyone who wants another way into the subject.
Future access
The plan is for EverythingAI to be free to use. It will not offer an accredited certificate or pretend to replace formal education. It is simply a place to learn differently, build understanding and become more comfortable with the language of AI.
EverythingAI organises learning into levels, modules and lessons with clear prerequisites. The roadmap shows what is available now, what has been completed and what still depends on earlier knowledge.
Each level can measure mastery from its lessons and review work. This means progress is based on completing the learning path, not simply opening pages.
A lesson can combine explanations, examples, activities, warnings, source references and practical checkpoints. The focused lesson workspace keeps the current material together and can reconnect it to earlier skills when the new idea depends on them.
The programme covers the practical, non-technical AI landscape and builds toward more advanced responsible-system work.
The current content includes design-preparation checkpoints across earlier levels and a larger advanced course for implementing and testing Sorelium Tower. Practical work remains locked until the required governance and safety learning has been completed.
This keeps “building with AI” connected to evidence, limits, testing and responsibility rather than treating implementation as the first lesson.
Assessments
Lessons can include different question types and completion requirements. Quiz results are checked against the lesson content, while written responses use guided self-review instead of pretending that an automated grade understands every personal answer.
Lessons and later levels can remain locked until their required knowledge or review work is complete.
Revision
The revision area can build review quizzes around a lesson, module or level. It uses progress and previous results to give learners another way to practise instead of forcing them to reread every page from the beginning.
Motivation
EverythingAI tracks experience and achievement progress from real learning actions. The achievement area works like a trophy case, showing what has been earned and how close the learner is to the next milestone.
Learners can save their own notes and bookmark useful lessons or sections, then return to those items from dedicated views instead of searching through the whole course again.
The glossary explains recurring terms, the sources area keeps supporting material visible, and the catalogue provides a wider view of the available learning content and its structure.
A command-style search can look across the installed learning package and take the learner back to matching lessons or reference material. It runs locally and does not require an external search service.
Local-first privacy
Progress, quiz attempts, notes, bookmarks and achievements are stored in a local database. The learning platform does not require a paid API, cloud database, account provider, external search service or model download for its core learning experience.
The current version is designed as a single-user local application, which is one reason it is not ready to be offered publicly yet.
Safe content updates
Learning content is stored in versioned packages with structural checks for lessons, links and prerequisites. A new package can be staged and validated before activation, and an earlier release can be restored without deleting the learner’s progress.
For a future public version, this provides a base for improving the course while keeping learning records separate from the content itself.
Start at the right level and follow lessons in a useful order.
Use clear lessons to understand what the tool or concept actually does.
Complete activities, assessments and revision to test understanding.
Use saved progress, notes, bookmarks and achievements to continue learning.
This is a preview of the idea. The project is still being developed privately and is not available for public use yet.