Learning and careers decision
Oncourse AI
A single developer can build a useful, narrow replacement (flashcards, spaced repetition, LLM tutor, PDF→flashcards) in multiple weeks, but reproducing the full commercial product — large curated medical content, polished native apps, and App Store distribution — is impractical without the vendor's content and user base.
View on the App Store↗Not priced
No pricing page we fetched carried a figure, so there is nothing to compare against. The build side is still real.
$100one-off140 h to build
$200/mo6 h/mo upkeep
No published price to break even against.
No open-source build does this yet
Nothing published replaces this one, so a replacement starts from an empty file. Here is what it would have to cover.
What a replacement has to do
- Recommend daily mix of questions, flashcards, notes and cases; present flashcards with spaced-repetition; show exam-style questions and explanations; accept PDF/pages to generate flashcards; provide an LLM-driven chat tutor that personalizes the daily plan.
What it still won’t have
- Curated, continuously-updated medical content and large question bank (40,000+ flashcards, 10,000+ questions)
- Polished native iOS UX and App Store distribution
- Developer-provided clinical case simulations and academy-style video lessons
- Existing user data, streak/engagement system and brand trust
What remains hard
- Product polish and ongoing maintenance
First-year cost
No published price
Oncourse AI does not publish a price we could read, so there is nothing to compare against. What building costs is below.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a minimal self-hosted study service: backend in Node.js (Express) with Postgres, a React web UI and an optional React Native iOS wrapper; integrate OpenAI-compatible LLM API for the AI tutor; implement Apple IAP subscription validation server-side (or Stripe for web users); core features: (1) import PDF/image pages and extract text to generate candidate flashcards, (2) store/manage flashcards and exam questions in Postgres, (3) spaced-repetition scheduler that produces a daily study plan, (4) LLM chat endpoint that gives concise high-yield notes, explanations, and recommends next items, (5) simple analytics (accuracy, topic mastery, time spent). Out of scope: curated medical Qbank content population, polished App Store listing, video lessons, large curated case library. Include input validation, retry/error handling for API calls, background worker for ingestion and scheduling (e.g. BullMQ), and automated tests for import flow, scheduler, and LLM integration.
How we checked
How the score was reached
- Partly verdict base52
- 3/3 assessment runs agreed+4
- Evidence score56
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.
How scoring works →Cited sources · 1
Every page the run actually retrieved.
- official productOncourse AI for USMLE - App Store
Integrity checks
What held up, and what did not.
