PROPELOO

DERIVATIVES EXCHANGE / PERPS & OPTIONS

Build a derivatives exchange where the risk engine never fails.

PROPELOO engineers cryptocurrency derivatives exchanges — perpetual futures, delivery futures, options and margin trading platforms. The defining challenge in derivatives is the risk engine: it must liquidate positions before they become insolvent, manage the insurance fund, calculate mark prices that cannot be manipulated, and handle cascading liquidations during market crashes without creating systemic bad debt.

A derivatives exchange that accumulates bad debt during a flash crash loses user trust it cannot recover.

The May 2021 crash and March 2020 flash crash demonstrated which derivatives exchanges had well-engineered risk systems and which did not. Exchanges with slow liquidation engines, poorly calibrated insurance funds, or mark prices that lagged the real market accumulated bad debt that was socialised to profitable traders — destroying the trust of their best users. A derivatives exchange risk engine must: liquidate at-risk positions before they go underwater (requires mark price that tracks index price accurately), maintain an insurance fund large enough to absorb typical market volatility, support partial liquidation (liquidating enough of a position to return to safe margin, not the whole position), and handle auto-deleveraging (ADL) transparently when insurance is exhausted.

The derivatives exchange stack.

System Layers

  • Risk Engine Layer: Margin monitoring, liquidation triggering, partial liquidation, insurance fund, ADL
  • Mark Price Layer: Index price construction, mark price calculation, funding rate calculation
  • Position Layer: Cross/isolated margin, position management, PnL calculation, leverage
  • Matching Layer: Perpetual order book, funding settlement, delivery settlement
  • Options Layer: Options pricing (Black-Scholes), Greeks, settlement, portfolio margin

Core Technical Capabilities

  • Liquidation & Risk Engine

    Real-time maintenance margin monitoring, automatic liquidation trigger, partial liquidation (reduce position to safe margin rather than full close), liquidation price display, take-over liquidation mechanism.

  • Funding Rate Mechanism

    Funding rate calculation from perpetual-spot premium, 8-hour settlement interval, interest rate component, funding rate display and history, impact of funding on position P&L.

  • Mark Price & Index

    Index price constructed from weighted average of multiple spot exchange prices. Mark price uses EMA of index to reduce manipulation. Liquidation based on mark price, not last traded price.

  • Insurance Fund

    Insurance fund accumulates from liquidations where collateral exceeds debt. Depleted when liquidations result in deficits. Fund size display, ADL trigger conditions, socialised loss mechanics.

  • Cross & Isolated Margin

    Cross margin: all collateral shared across positions, higher capital efficiency. Isolated margin: separate collateral per position, maximum loss limited to isolated margin. User-selectable per position.

  • Options Trading

    Vanilla options (calls/puts), European/American settlement, Black-Scholes pricing with IV surface, Greeks display (delta, gamma, theta, vega), portfolio margin for delta-neutral strategies.

How we think about derivatives exchange design.

Risk management in a derivatives exchange is the product. The trading interface is packaging.

  • Mark price manipulation resistance is non-negotiable

    Exchanges that liquidate based on the last traded price (which can be manipulated with a small order in a thin market) will liquidate users incorrectly during low-liquidity periods. Mark price based on a time-weighted average of a multi-source index price eliminates this attack vector and is the industry standard.

    Axiom:

  • Insurance fund calibration is ongoing, not a launch parameter

    The insurance fund must be large enough to absorb the worst realistic market event without triggering ADL. Initial fund sizing based on historical volatility, open interest and liquidation engine speed. Monitor fund drawdown events and adjust minimum fund target after each stress event.

    Axiom:

  • Partial liquidation reduces cascades

    Full liquidation (close the entire position when margin is insufficient) amplifies market volatility by creating large sell orders. Partial liquidation (close enough of the position to return the account to the maintenance margin level) is gentler on the market and better for the liquidated user. It also reduces the probability of triggering cascading liquidations in correlated positions.

    Axiom:

  • Options require different risk management

    Options have non-linear payoffs — gamma risk creates dramatically different risk profiles near expiry. Options portfolio margin (netting delta, gamma and vega across a portfolio of options positions) is more capital-efficient and realistic than position-by-position margin but requires a full options risk model. Most exchange options implementations support basic scenario margin initially.

    Axiom:

Derivatives exchange design decisions.

  • On-chain vs off-chain order book?

    Impact: Off-chain matching + on-chain settlement is the practical production architecture for most derivatives DEXes — matching engine speed with trustless fund custody.

    • On-chain order book — fully decentralised, expensive gas, slow
    • Off-chain order book, on-chain settlement (dYdX v3 style) — fast matching, trustless settlement
    • Centralised matching, on-chain custody (most CEX) — fastest, counterparty risk
    • Hybrid (off-chain matching + ZK proof settlement) — dYdX v4, most complex
  • Perpetuals vs delivery futures?

    Impact: Perpetuals at launch — they account for 90%+ of crypto derivatives volume. Add delivery futures for institutional clients and options as a premium product.

    • Perpetuals only — no expiry, highest volume, funding rate mechanics
    • Delivery futures only — institutional, predictable settlement date
    • Both — full product suite, higher complexity
    • Options only — specialised, sophisticated users
  • Cross vs isolated margin default?

    Impact: User-selectable per position (cross or isolated) is the industry standard. Default to isolated for new users (limits maximum loss). Power users prefer cross for capital efficiency.

    • Cross only — simpler, capital efficient, user confusion risk
    • Isolated only — safer for users, less capital efficient
    • User selects per position — standard in production exchanges
    • Portfolio margin (netting) — most sophisticated, complex to implement
  • Insurance fund model?

    Impact: Dynamic insurance fund that receives a percentage of profitable liquidations is the sustainable model. Fund grows with the exchange's open interest.

    • No insurance fund — all bad debt via ADL
    • Fixed insurance fund — simple, may run out
    • Dynamic (% of liquidation proceeds) — grows with usage
    • Hybrid (fixed floor + dynamic contribution) — most robust
  • Mark price oracle?

    Impact: Pyth Network for low-latency mark price construction (critical for accurate liquidation), with circuit breakers that pause liquidations when Pyth price deviates from other sources by more than a threshold.

    • Self-constructed (weighted average of spot exchanges) — no external dependency
    • Chainlink price feed — manipulation resistant, some pairs not available
    • Pyth Network — low latency, pull-based, good derivative exchange support
    • Hybrid (primary + fallback + circuit breaker) — most robust

What PROPELOO builds.

  • Perpetual Futures Exchange

    Perpetuals exchange with funding rate mechanism, cross/isolated margin, partial liquidation, insurance fund, mark price oracle and professional trading UI.

  • Options Exchange

    Vanilla options trading with Black-Scholes pricing, IV surface, Greeks display, European settlement and portfolio margin for institutional traders.

  • Margin Trading Platform

    Spot margin trading with lending pool, configurable leverage, automatic margin calls, collateral management and lending rate analytics.

  • Copy Trading Exchange

    Social/copy trading platform built on top of derivatives — performance leaderboard, one-click copy, position mirroring with configurable size multiplier.

  • Derivatives Risk System Rebuild

    Replace or upgrade risk engine for existing exchange — faster liquidation, improved mark price, insurance fund modelling, ADL transparency.

  • Institutional Derivatives

    FIX protocol derivatives exchange for institutional clients — prime brokerage, portfolio margin, block trading, OTC desk integration.

The derivatives exchange stack.

  • Matching & Risk

    Stack: Go / Rust (engine), Redis (real-time positions), Kafka (event log), Custom risk engine

  • Oracle

    Stack: Pyth Network, Chainlink Data Feeds, Custom multi-source index, TWAP calculation

  • Database

    Stack: PostgreSQL (settlement), TimescaleDB (price data), Redis (positions/margins)

  • Options

    Stack: QuantLib (options pricing), Black-Scholes (IV surface), Custom Greeks calculator

  • Frontend

    Stack: React + TradingView, Real-time P&L display, WebSocket (positions), Mobile (React Native)

  • On-chain (DEX)

    Stack: Solidity / Rust, Foundry + Echidna, Off-chain matching, ZK settlement (advanced)

Derivatives exchange security includes economic attack vectors.

  • Mark price manipulation

    Any mechanism that can move the mark price can trigger incorrect liquidations. Multi-source index with outlier rejection, TWAP smoothing and circuit breakers that halt liquidations on anomalous mark price movement.

  • Liquidation front-running

    Liquidators front-run each other to capture liquidation bonuses. Fair liquidation bonus (large enough to incentivise immediate liquidation, small enough not to create adverse incentives) with priority-ordering for largest-risk positions.

  • Insurance fund draining

    Coordinated market manipulation that forces many liquidations simultaneously, depleting the insurance fund. Fund monitoring, maximum insurance drawdown alerts and market circuit breakers.

  • Oracle price manipulation

    A compromised oracle that reports a manipulated mark price can trigger mass liquidations. Multi-source oracle with deviation detection, circuit breakers on oracle price anomalies, fallback oracle configuration.

  • Matching engine integrity

    Immutable event log of all order events. Reconciliation between matching engine state and on-chain settlement. Replay capability for recovery from engine failure.

  • Key management for settlement

    For on-chain settlement: multi-sig or MPC key management for settlement signing. HSM-backed signing for large settlement batches. No single point of key failure.

Frequently Asked Questions

What is a funding rate and why is it necessary?

Perpetual futures have no expiry, so without a correction mechanism, their price would diverge from the spot price permanently. The funding rate is the correction mechanism: when perps trade above spot, longs pay shorts periodically (discouraging longs, encouraging shorts, pulling the perp price down toward spot). When perps trade below spot, shorts pay longs. The rate is proportional to the premium of the perp over the spot index. 8-hour settlement is the industry standard.

What is auto-deleveraging (ADL)?

When the insurance fund is depleted and a liquidation results in a deficit, the exchange must cover the loss somehow. ADL automatically reduces the positions of the most profitable traders on the opposite side proportionally to cover the deficit. ADL is a last resort — it takes profit from traders who did nothing wrong. Most well-run exchanges rarely trigger ADL because their insurance funds are large enough. ADL events are a red flag that the insurance fund is undersized or the risk engine is too slow.

What is the difference between cross and isolated margin?

Isolated margin: the position has its own dedicated collateral. If the position is liquidated, only that collateral is lost — other positions are unaffected. Maximum loss is the isolated margin amount. Cross margin: all available account balance is shared across all open positions as collateral. Higher capital efficiency (one position's unrealised profit can cover another's margin call) but maximum loss is the entire account balance.

How does the mark price oracle prevent market manipulation and flash liquidations?

We construct mark prices using a volume-weighted spot index from multiple external tier-1 exchanges (Binance, OKX, Kraken, Coinbase) combined with an exponential moving average (EMA) of the perp basis. Outlier filtering and median calculation strip anomalies. Because positions are evaluated against this composite mark price rather than volatile local order book trades, malicious wick manipulation or thin order book spoofing cannot trigger unwarranted liquidations.

What order matching throughput and latency are required for derivatives trading?

Derivatives exchanges require deterministic sub-millisecond execution because rapid order cancellations and market-maker quote updates generate 10x higher load than spot exchanges. Our matching engine written in Go and Rust handles 100,000+ orders per second per market pair with p99 latency under 800 microseconds, backed by memory-mapped order books and sequenced event sourcing.

How does partial liquidation reduce market impact compared to full liquidation?

Full liquidation dumps 100% of an underwater trader's position into the order book at market, frequently creating massive slippage and triggering cascading liquidations. Our tiered partial liquidation engine liquidates only enough tranches to return the account to safe maintenance margin levels, reducing market impact and preserving remaining trader equity whenever possible.

How do you handle options pricing and portfolio risk (Greeks and Black-Scholes)?

For crypto options platforms, we integrate real-time Black-Scholes and jump-diffusion pricing models with dynamic implied volatility (IV) surfaces. The risk engine continuously computes portfolio Greeks (Delta, Gamma, Vega, Theta, Rho) and supports cross-asset portfolio margining, allowing hedged strategies (such as delta-neutral straddles or covered calls) to benefit from reduced margin requirements.

How are cryptocurrency derivatives settled (cash settlement vs physical delivery)?

Most crypto perpetuals and dated futures settle in cash (USDT/USDC margined or Coin-margined in BTC/ETH). For inverse contracts, collateral and profit/loss are denominated in the underlying cryptocurrency. For linear contracts, settlement is in stablecoins. In dated futures, expiring contracts are automatically cash-settled against the 30-minute TWAP index price at expiry.

Can institutional market makers connect via FIX protocol or high-frequency WebSocket APIs?

Yes. Our derivatives exchange architecture includes low-latency FIX 4.4/5.0 protocol gateways and binary WebSocket streaming feeds. Market makers receive dedicated rate-limit allocations, co-location options, mass-quote cancellation endpoints, and sub-millisecond execution reports to maintain tight spreads.

What regulatory compliance tools are embedded into the derivatives exchange platform?

The platform integrates automated Travel Rule messaging (Notabene/TRISA), real-time on-chain transaction monitoring (Elliptic/Chainalysis) for deposits and withdrawals, automated leverage capping based on user jurisdiction, and automated AML/CFT risk scoring to satisfy strict global derivatives licensing requirements.