Wondering

The centerpiece is the course generator. You give Wondering a topic and some context about yourself, and it returns a course outline with a deliberate order. The onboarding asks what you do and what you want to learn, and the platform says it uses examples relevant to your role and expertise. The pitch is that two people asking for a course on the same subject get different courses, because they have different jobs. Why does that matter? Because most learning platforms treat everyone the same. You start at lesson one and wade through fundamentals you already know before reaching anything useful. Wondering tries to cut that front-loaded boredom. You keep what you care about, explore other angles later, and the outline is editable before you start. The free tier lets you test the generation flow before paying, which is the right way to do it. Generate a course, read the first outline, and decide whether the structure actually matches what you wanted.

Lessons are sized for real life. The platform talks about three minutes a day, lessons a commute long, and a version that fits in the gap between meetings. Each lesson blends visuals with text and audio rather than dumping a transcript on you. The claim is a full course of short, three-minute audio and visual lessons personalized to you. The visual approach is the part that stands out. Instead of a chat window that spits out paragraphs, Wondering renders explanations as structured visual content. That's a real difference if you've ever tried to learn from a wall of AI-generated prose and found your eyes glazing over. Short lessons also lower the activation energy. A three-minute commitment is easy to say yes to, and the platform nudges you toward doing a little every day instead of bingeing and forgetting.

Every course can be consumed as audio. Wondering generates podcast-style lessons so you can learn while walking, driving, or doing dishes. On the Max tier, the platform lets you generate up to 90 podcasts a month, which is far more than most people would ever listen to. The audio mode is what makes the three-minute framing useful in the real world. Plenty of people can't sit still and read a lesson, but they can listen to one on the way to work. It's not a replacement for the visual lessons. It's the same material in a format that fits a different part of your day. If you learn best by listening, this is the feature that decides whether Wondering works for you.

Wondering doesn't stop at consumption. Courses include practice, described as multiple formats covering flashcards, multiple choice, and cloze deletion. Cloze deletion is the fill-in-the-blank style that forces active recall instead of passive recognition, and it's a common technique in serious learning tools. The platform pairs practice with spaced repetition, resurfacing material you're about to forget. This is the mechanism that separates learning from reading. You can watch a hundred videos and retain almost nothing. You can answer a handful of well-timed questions and retain far more. Wondering tracks a split between what you've learned and what still needs review, so you know where to spend your attention instead of guessing.

Beyond static lessons, Wondering offers live voice sessions with an AI tutor. You can ask the tutor a question and talk through a concept out loud. The free tier includes one live voice session a month, Pro bumps that to fifteen, and Max goes to forty-five. Talking through a concept is one of the fastest ways to find out whether you actually understand it. The voice tutor gives you something to think against, which is more useful than rereading a lesson when you're stuck. The monthly caps are the honest limitation here. Fifteen sessions a month is plenty for most people, but one on the free tier is a taste, not a habit. If live tutoring is the feature you came for, factor the tier limits into your decision.

One of Wondering's more unusual features is an MCP server that connects the platform to AI assistants like Claude Desktop or Cursor. That means you can ask your assistant to pull in your learning context, or to have Wondering generate a course and quiz you on it. The platform even ships example prompts, like asking your assistant to find what you learned yesterday and quiz you using the practice questions from those lessons. This is available on Pro and Max, not the free tier. It's the feature that turns Wondering from a standalone app into part of a larger workflow. If you already live in Claude or Cursor, having your learning platform speak the same protocol is genuinely useful. For everyone else, it's a nice extra rather than a reason to subscribe.

Underneath the generation, Wondering points to a reference library built on openly licensed material. The platform mentions more than 1,500 openly licensed textbooks and a workflow that turns them into personalized courses. It also describes courses built from real research, such as studying design roles at OpenAI, Anthropic, Google DeepMind, Cursor, and Figma to map what AI now asks of designers. That sourcing approach matters for trust. An AI course is only as good as what it was built from, and citing real material is better than generating plausible-sounding filler. It doesn't guarantee every lesson is accurate, and you should still treat AI-generated content as a starting point rather than gospel, but grounding courses in identifiable sources is the right instinct.