Skip to content
Finance Lessons

DeFi Lending & Borrowing

Final Exam: The DeFi Lending Gauntlet

A graded, locked capstone exam — 26 one-shot questions spanning every DeFi lending lesson: pooled money markets, collateral and LTV, the interest-rate model and utilization, the health factor and liquidations, and flash loans, oracles and the real risks.

16 min Updated Jun 3, 2026

This is the part where the safety nets come down. No pretests, no warm-up analogies, no “guess before reading” — just the course, asking whether any of it stuck. Everything you need, the previous five lessons already taught you. Read each option twice; the trap is usually the one that’s 90% right.

Warning:

How this exam works

This is a graded exam. Questions come one at a time. Once you submit an answer it is final — there is no going back, no second try, and a wrong answer simply fails that question. Your score stays hidden until the end, where you need 70% to pass. Read every option twice before you commit.

Question 1 of 26

A friend says 'a DeFi lending protocol is just a bank that lives on a blockchain.' What is the sharpest correction?

Select an answer to continue.

Course Recap

Big picture

The whole DeFi lending machine

  • DeFi lending, end to end
    • Pooled money markets
      • Shared contract pool, not a bank manager
      • Suppliers earn, borrowers draw, same pool
      • Receipt token (aToken/cToken) accrues yield
      • Permissionless, no credit check, no KYC
      • Over-collateralized: no recourse, no identity
    • Collateral & LTV
      • LTV = debt ÷ collateral
      • Max LTV caps borrowing; liq. threshold is higher
      • Gap between them is the safety buffer
      • Volatile assets get lower collateral factors
      • Borrowing power = collateral × max LTV
    • Interest-rate model
      • Utilization U = borrowed ÷ supplied
      • Kinked curve: gentle below optimal, steep above
      • Steep zone defends withdrawable liquidity
      • Supply rate = borrow × U × (1 − reserve factor)
      • Supply rate below borrow rate, always
    • Health factor & liquidation
      • HF = Σ(collateral × liq. threshold) ÷ debt
      • HF below 1 means liquidatable
      • Keepers/bots liquidate for the bonus
      • Liquidation penalty pays the liquidator
      • Close factor caps debt repaid per liquidation (≤50%)
    • Flash loans & risks
      • Flash loan: uncollateralized because atomic
      • Repaid in one tx or it all reverts
      • Flash-loan attacks exploit oracles, not the loan
      • Bad debt → reserves & safety module backstops
      • E-mode, isolation mode, looping shift risk
Five lessons, one structure: the pooled money market that replaces the bank, the collateral and LTV that bound a loan, the utilization curve that prices it, the health factor that polices it, and the flash loans, oracles and risks that surround it.

Key Takeaways

Success:

What to remember

  • A money market is a pooled, autonomous contract, not a bank — suppliers and borrowers share one liquidity pool, supplying mints an interest-bearing receipt (aToken/cToken), it’s permissionless with no credit check, and loans are over-collateralized because the protocol has no identity or recourse to fall back on (Aave, Compound, MakerDAO).
  • LTV bounds the loan — LTV = debt ÷ collateral; the max LTV caps how much you can borrow while the higher liquidation threshold is where you’re liquidated, and the gap between them is your buffer. Volatile assets get lower collateral factors, and borrowing power = collateral × max LTV.
  • Utilization prices borrowing — U = borrowed ÷ supplied drives a kinked two-slope curve (gentle below the optimal point, steep above it to defend liquidity). The supply rate ≈ borrow rate × U × (1 − reserve factor), so suppliers always earn less than borrowers pay.
  • The health factor polices solvency — HF = Σ(collateral × liquidation threshold) ÷ debt; once it drops below 1, keeper bots liquidate the position, repaying up to the close factor (≤50%) of the debt in exchange for collateral plus a liquidation bonus.
  • Flash loans and oracles define the real risks — a flash loan is uncollateralized only because it’s atomic (repaid in one transaction or the whole thing reverts); flash-loan attacks exploit a separate weakness, usually a manipulable oracle. Bad debt is backstopped by reserves and a safety module, and e-mode, isolation mode, and recursive looping all reshape the risk you take.

Mark lesson as complete