Ai Coding And Product Systems medium
You are launching a free tier for a chat assistant. Free users are projected to outnumber paid users 10 to 1, and finance requires free-tier inference cost per user to be a small fraction of paid-tier cost. Which design meets this?
- A Give free users a smaller, cheaper model with hard daily token budgets, and reserve the large model for the paid tiers.
- B Give free users the same model and limits as paid, relying on the fact that most free users send only a few messages per day.
- C Give free users the same model but lower request priority, so they wait in a longer queue during peak hours and costs stay flat.
- D Give free users the large model with a soft monthly budget that only shows an upgrade prompt once their usage passes the limit.
Answer: A. Give free users a smaller, cheaper model with hard daily token budgets, and reserve the large model for the paid tiers.
Free tiers need **hard token budgets and a degraded model**, trading answer quality for bounded cost per user. Lower priority (C) is tempting because it feels like rationing, but every token is still generated on the expensive model, so cost per user does not drop. B and D leave the heavy-tail users who drive most of the spend uncapped. Offering the paid model is reasonable only as a short, hard-capped trial meant to drive conversion.
One system design question a day, with spaced repetition so the ones you miss come back.
Get the appRelated questions
- Your assistant uses a classifier to route queries to a small or large model, which saves 60% of GPU spend. Analysis shows about 20% of hard queries are misrouted to the small model, which then gives confident wrong answers, and regenerate clicks on those threads have doubled. What should you do?
- Your workspace AI retrieves the top-k chunks from a shared vector index and passes them to the model with ACL metadata and a system prompt saying 'only use documents this user can access.' A user just received an answer quoting a private HR page they cannot open. What is the correct fix?
- Your support bot answers billing questions. One user has asked the same refund question three times, and retrieval confidence for that topic is low. What should the bot do next?
- Your AI review system receives a 4,000-line pull request spanning 60 files. Sending the whole diff in one prompt yields shallow, generic comments, and real bugs in files near the middle of the diff are missed. How should you structure the review?
- In your ChatGPT-style assistant, power users keep single threads running for weeks, reaching about 150K tokens. Cost per turn keeps climbing, latency is rising, and the model increasingly misses facts from early in the thread. What should you do?
- Your workspace AI filters retrieval using ACLs stored in the vector index, which is rebuilt nightly. An admin removes an employee from a confidential project page, but for the rest of the day the employee can still get AI answers drawn from it. What is the right fix?