Senior Quant Risk Engineer / Architect (Perpetuals & Liquidation Engine)

Overview：

As a Senior Quant Risk Engineer, you will hold direct accountability for the protocol’s Global Solvency. Your mission is to architect, optimize, and stress-test the core risk engine, covering dynamic tiered MMR/IMR matrices, multi-collateral cross-margin frameworks, and high-frequency liquidation matching logic. You will ensure that during black-swan events—marked by extreme market gapping, flash crashes, and sudden liquidity dry-ups—the system’s insolvency and bad-debt rate remains strictly at 0%.

Job Responsibilities：

1. Multi-Collateral Cross Margin & Dynamic Risk Parameter Design

• High-Dimensional Tiered MMR/IMR Models: Move beyond static percentages. Design and implement a dynamic matrix where Maintenance Margin Requirement (MMR) and Initial Margin Requirement (IMR) are functions of Notional Position Size, Live Orderbook Depth/Slippage Curves, and Implied/Realized Volatility. Ensure margin curves scale adaptively to cover market impact costs.

• Collateral Haircut Matrix Optimization: Define risk parameters for multi-asset collateral under a unified cross-margin framework. Integrate cross-asset correlation coefficients to calculate adaptive haircut factors, preventing users from borrowing premium assets against illiquid, un-clearant "long-tail" tokens.

• Anti-Manipulation Mark Price Mechanisms: Architect manipulation-resistant index and mark pricing algorithms. Combine Multi-source Oracle Medianizers, Time-Weighted Average Prices (TWAP), and Basis Correction (via Funding Rates) to bulletproof the system against flash-loan attacks and malicious predatory liquidations.

1. Liquidation Engine Optimization & Micro-Structure Mechanics

• Incremental & Partial Liquidation State Machine: Refine the execution flow when an account’s equity breaches the MMR threshold. Avoid "one-click total liquidations." Implement a step-by-step partial liquidation routine (e.g., liquidating 10-25% of the position per step) to bring the account back to the IMR safe line while dynamically scaling cooldown periods based on market volatility.

• Dutch Auctions & Keepers Incentive Optimization: Minimize the Market Impact Cost of liquidated positions. Engineer a Dutch auction clearing mechanism featuring dynamic slippage discounts to incentivize external Liquidators/Keepers to absorb toxic flow rapidly before prices breach the Bankruptcy Price.

• Auto-Deleveraging (ADL) & Global Solvency Backstops: Quantify the optimal capital injection and drawdown thresholds for the protocol’s Insurance Fund. Define the structural triggers and execution order for ADL under terminal liquidity scenarios, ensuring systematic bad debt is neutralized based on counterparty profit and leverage rankings.

1. Quant Simulator Construction & Extreme Stress Testing

• Agent-Based Liquidation Simulator: Build a high-fidelity, micro-structure market simulator using Python (NumPy, Pandas, SciPy). Model heterogeneous market participants (e.g., high-frequency arbitrageurs, panic-sellers, MEV bots, liquidators) to simulate historical crunches (e.g., March 12, FTX collapse) and quantify VaR (Value at Risk) and Expected Shortfall (ES).

Job Requirements：

1. Hardcore Industry Track Record (Strict Non-Negotiable)

• Professional Experience: 2+ years of hands-on quant risk or liquidation architecture experience at a Tier-1 CEX (e.g., Binance, OKX, Bybit Risk/Liquidation teams) or a premier decentralized perpetual protocol (e.g., dYdX, GMX, Synthetix, Hyperliquid).

• Core Mechanics Mastery: Flawless mathematical and structural understanding of Mark Price, Liquidation Price, Bankruptcy Price, Dynamic Funding Rates, Portfolio/Cross Margin, and the full lifecycle of a derivatives order. You must be able to spot edge-case mathematical exploits in formulas at a glance.

1. Tools & Engineering Capability

• Tech Stack: Master of Python for quantitative analysis, modeling, and backtesting. Proficiency in SQL/ClickHouse to parse and extract statistical features from tens of millions of rows of high-frequency orderbook and chain data.

• Smart Contract Literacy: While full-time smart contract engineering is not required, you must be able to audit and read Solidity/Rust code with ease. You will ensure that your quantitative risk models translate into smart contracts without floating-point precision loss, integer overflows, or race conditions caused by on-chain latency.

1. Execution & Communication Style

• Highly results-oriented. Ability to translate abstract stochastic processes or high-dimensional risk matrices into clear algorithm pseudocode, If-Else conditional state machines, and rigorous edge-case test suites for core smart contract developers.