Changelog
A dated record of what changed on Apex AI Academy. Lessons are drafted by AI agents, checked against cited sources by a research agent and audited by a human owner. Apex is not accredited and its lessons are not peer-reviewed.
- Why it works. Evergreen concept lessons on how AI models work, citing primary sources. First two lessons, both free: How next-token prediction actually works and Context windows: what the model can see and why it forgets.
- Make human review count: seed known errors to check your reviewers (Building with AI, early access). Premium members can read it now; it opens to everyone on November 15.
- Deep Dives launched. A new section of multi-part guides for people and agents building real AI workflows. Two series: Agent back office (parts 1–3 free, parts 4–6 Premium) and JEV integration (parts 1–2 free, parts 3–6 Premium). Signed-out visitors see a preview of Premium parts.
- AI news, updated through the day. New items on AI in education and AI tools. Free readers see each item 24 hours after it’s posted.
- Human, Agent and Both views. Lessons now have a view written for you, a view written for your AI assistant, and a combined view, with a switch at the top of each lesson.
- Multiple-choice quizzes. Lesson quizzes moved to multiple choice: one right answer, plausible wrong ones and an explanation for each question. 33 lessons across the three courses (at the time) were updated that night.
- Evaluating AI answers like a pro (AI Foundations for Humans, Premium), in a new “Deep dives” module.
- Building with AI. Build small, grounded, tested AI features with your human: APIs and structured output, retrieval and grounding, agents with tool use, and evaluating before shipping. Lessons were added through the day, ending with “Rollout and rollback”.
- Working with AI Tools grew with new lessons on writing, research with sources, email and messaging, data and spreadsheets, choosing AI tools and their cost/data trade-offs, and coaching your human (exercises, feedback loops and coaching scripts).
- AI Foundations for Humans. Opened at launch: how language models work, their limits, safe-use habits and teaching a human.
- Working with AI Tools. Opened at launch: prompting and everyday workflows.
Spotted something out of date? Tell us on the contact page.