The core risk of digital asset derivatives business lies not in matching engine performance, but in the reliability of the liquidation system. Under Malaysia's RMO DAX framework, derivatives exchanges must demonstrate that their liquidation mechanisms can effectively liquidate risky positions during extreme market conditions, protecting the exchange, insurance fund and retail users from bankruptcy losses. For platforms serving local investors and institutional clients, a production-proven liquidation engine is a prerequisite for license retention, institutional onboarding and long-term operations.

But a truly engineered liquidation system is far more complex than "force close when account loss reaches a certain ratio." It must answer a series of questions:

• Should the margin model be isolated, cross, or hybrid? How should initial margin rate and maintenance margin rate be set to control platform risk without hurting capital efficiency?
• Should liquidation triggers use fixed thresholds or staged triggers? Should liquidation orders be sent to the public market or first taken by market makers? Should small positions be liquidated before large ones, or vice versa?
• If liquidation orders cannot be filled and cause bankruptcy losses, should the insurance fund absorb losses first, or should ADL auto-deleveraging be triggered immediately to have profitable users share losses proportionally?
• How should the insurance fund be raised, used and replenished? What are the SC requirements for segregating platform capital and client funds?
• How to ensure the availability of the liquidation engine itself? What is the degradation process if the engine goes down during extreme volatility?

This article explains the complete architecture of a derivatives liquidation risk engine for compliant Malaysian exchanges from both product and engineering perspectives, including margin models, trigger mechanisms, liquidation execution, ADL, insurance funds, clawback rules, bankruptcy price calculation, regulatory compliance and audit trails. It also presents SoonTech's proven solutions from multiple Southeast Asian exchange projects. Whether you are a product team building a derivatives exchange, a technical lead responsible for risk control, or an operator applying for an RMO DAX license, you will get a complete blueprint that can be directly referenced for design and review.

1. Industry Background: Why Liquidation Risk Is a Core SC Focus

1.1 Explicit RMO DAX Requirements for Derivatives Risk

The Securities Commission Malaysia explicitly requires in the RMO DAX guidelines that licensed digital asset exchanges must establish a sound risk management framework including margin systems, liquidation mechanisms, position limits and stress testing. Platforms must demonstrate:

• Liquidation trigger rules are clear, transparent and applied uniformly to all users;
• The system can complete liquidation without systemic bankruptcy under extreme market conditions;
• User funds and platform funds are fully segregated, and the insurance fund is not the operator's property;
• All liquidation events have complete audit records, and users have the right to appeal and review;
• Platforms regularly conduct stress tests simulating liquidation capacity under 20%+ single-day price movements.

These are not paper compliance requirements—SC will request liquidation system design documents, historical liquidation data analysis and stress test reports during annual license reviews and on-site inspections.

1.2 Special Characteristics of the Southeast Asian Derivatives Market

Compared to mature markets, the derivatives markets in Malaysia and Southeast Asia have several distinctive features:

• User structure is retail-heavy, with a low proportion of professional investors and insufficient understanding of margin, liquidation and bankruptcy risks;
• Correlation between local currencies and major coins is higher, so black swan events in a single currency easily spread to the entire market;
• Local market maker depth is limited, and liquidity may disappear instantaneously during extreme volatility;
• Retail users are accustomed to high leverage (10x-100x), further amplifying liquidation pressure.

This means Malaysian exchanges cannot directly copy liquidation parameters from Binance or OKX—they must be localized according to local market depth, user structure and regulatory requirements.

1.3 Precedents of Liquidation Accidents

Liquidation system failures causing accidents are not uncommon globally:

• One exchange batch-liquidated users who should not have been liquidated due to a staged liquidation logic bug during severe price volatility;
• One platform triggered ADL after insurance fund depletion, and profitable users had over 30% of their profits clawed back, triggering massive customer complaints and regulatory intervention;
• One exchange's liquidation engine went down for 40 minutes during high concurrency, resulting in over $100 million in bankruptcy losses eventually covered by shareholders;
• One platform was fined millions of dollars by regulators for its "liquidate first, notify later" design deemed inadequate information disclosure.

For licensed Malaysian platforms, a serious liquidation accident means not only financial losses but potentially license suspension or revocation.

2. Margin Models: Isolated vs Cross vs Hybrid

2.1 Isolated Margin

In isolated mode, each contract position occupies an independent margin account, and P&L is settled only within that account without sharing with other positions. Characteristics:

• Risk isolation: Liquidation of one contract does not affect other positions or account balance;
• Intuitive user experience: Users can clearly see how much margin each contract uses;
• Lower capital efficiency: Users must deposit margin separately for each contract;
• Simple liquidation logic: Independent calculation per contract with no cross-impact.

Isolated margin is suitable for retail users and beginners, and is also the default mode recommended by Malaysian regulators—SC explicitly requires platforms to provide isolated margin as an option for retail users in investor protection guidelines.

2.2 Cross Margin

In cross margin mode, all available balance in the account serves as common margin for all positions, and P&L is calculated across the entire account. Characteristics:

• High capital efficiency: One margin deposit supports multiple contracts;
• Risk diversification: Floating profits from winning positions can offset floating losses from losing positions;
• Complex liquidation logic: Real-time consolidated calculation of full-account risk exposure is required;
• Risk contagion: Liquidation of one contract may consume the entire account balance.

Cross margin is suitable for professional traders, market makers and institutional clients. Malaysian platforms typically only open cross margin to clients who have completed professional investor certification and require signing an additional risk disclosure statement.

2.3 Hybrid Mode and Partial Margin Isolation

Major platforms adopt hybrid modes: Users can select isolated or cross margin per contract, and within cross margin accounts, "isolated partial margin" sub-accounts can be set. SoonTech's implementation experience in Malaysian projects:

• Retail users default to isolated margin; cross margin requires manual activation with pop-up warnings;
• Professional users can customize "cross groups," putting related contracts (e.g., BTC perpetual + BTC quarterly) into the same margin pool;
• Platform-wide "maximum cross leverage cap" (e.g., max 20x cross, max 50x isolated);
• Cross margin mode must display three core metrics in real time: current margin ratio, liquidation trigger price, estimated bankruptcy loss.

2.4 Initial Margin Rate and Maintenance Margin Rate

The two most core parameters:

• Initial Margin (IM): Minimum margin ratio required when opening a position, determining maximum leverage. For example, IM = 1% corresponds to max 100x leverage.
• Maintenance Margin (MM): Minimum margin ratio that must be maintained during holding; liquidation is triggered when below this value. MM is usually lower than IM, e.g., IM = 1%, MM = 0.5%.

Malaysian regulators tend toward "lower leverage, lower risk." SC recommends maximum leverage not exceeding 20x for retail users and 50x for professional investors. SoonTech's recommended configuration for local projects:

3. Liquidation Trigger Mechanism: From Fixed Threshold to Staged Liquidation

3.1 Three Liquidation Trigger Modes

Major exchanges use three trigger modes:

1. Fixed threshold trigger: When margin ratio ≤ MM, liquidate the entire position at once. Simple logic but brutal experience—minor price fluctuations can liquidate everything.
2. Staged trigger: When margin ratio drops from 2% → 1.5% → 1% → 0.8% → 0.5%, only liquidate a portion of the position each time, leaving time windows for users to add margin. Better experience but more complex implementation.
3. Incremental trigger: Calculate distance from current price to liquidation price in real time, send advance warning notifications, and add additional notifications every 10% the price moves unfavorably.

Malaysian regulators prefer staged triggers because they maximize retail user protection. SC explicitly recommends in investor protection guidelines that platforms "adopt progressive liquidation and avoid liquidating entire positions at once."

3.2 Engineering Implementation of Staged Liquidation

A typical 5-stage liquidation flow:

• Warning level (margin ratio 150% MM): Push App, email, Telegram notifications reminding users to add margin or reduce positions;
• Level 1 liquidation (120% MM): Liquidate 20% of position, send notification;
• Level 2 liquidation (110% MM): Liquidate another 30% (50% total), send notification;
• Level 3 liquidation (100% MM): Liquidate another 30% (80% total), send notification;
• Final liquidation (≤ MM): Liquidate remaining 20% of position.

Key design points:

• After each stage, leave a 30–60 second window for users to add margin before continuing;
• If users add margin and margin ratio rises back to the previous level, stop subsequent liquidation;
• If price moves favorably, also stop subsequent liquidation;
• Every liquidation action must have a complete audit log including trigger time, trigger price, liquidated quantity, execution price, actual P&L.

3.3 Liquidation Priority Sorting

When the system has many liquidation orders to process simultaneously, the order of liquidation directly affects overall efficiency and bankruptcy risk. Common sorting strategies:

• Lowest margin ratio first: Liquidate the most dangerous positions first to prevent bankruptcy from expanding;
• Smallest position first: Liquidate small orders first—fast execution prevents small orders accumulating into large bankruptcy losses;
• Largest position first: Liquidate large orders first to reduce overall system risk exposure;
• Highest liquidity first: Liquidate contracts with deepest order books first for easier execution and less market impact.

SoonTech uses a hybrid strategy in Malaysian projects: First sort by margin ratio tiers, within the same tier liquidate small orders before large ones, and during extreme volatility pause large order liquidation or split them into smaller iceberg orders.

3.4 Market Handling of Liquidation Orders

Liquidation orders should not be sent directly to the public order book—otherwise bots will front-run them, causing worse slippage and more bankruptcy. The correct processing flow:

1. Market maker priority uptake: Send liquidation orders to whitelisted market maker dark pools first, giving a 50–100 ms priority execution window;
2. Internal matching: Check if opposite-direction limit orders exist in the user order book for internal matching;
3. Public market execution: Split unfilled portions into small orders and gradually send them to the public market;
4. ADL fallback: If all above fail, trigger auto-deleveraging.

This three-tier architecture—dark pool first, internal matching, public market fallback—minimizes liquidation market impact and significantly reduces bankruptcy probability.

4. ADL Auto-Deleveraging Mechanism Design

4.1 What Is ADL

ADL (Auto-Deleveraging) is a safety mechanism: When liquidation orders cannot be filled in the market and the insurance fund is insufficient to cover bankruptcy losses, the system automatically reduces positions of profitable users in order of profit percentage, using their profits to fill the bankruptcy gap.

For exchanges, ADL is the final safety net—it ensures that under any extreme market conditions, the platform itself does not need to use its own capital as a backstop, nor does a single user's bankruptcy create systemic risk.

4.2 ADL Sorting Rules

The core of ADL is "whose position to reduce first." Common sorting dimensions:

• Profit percentage priority: Most profitable users are deleveraged first—they have the strongest risk-bearing capacity and should most be responsible for extreme market conditions;
• Leverage priority: High-leverage users are deleveraged first—they enjoyed the benefits of high leverage and should bear corresponding risks;
• Position age priority: Short-term speculators are deleveraged before long-term hedgers;
• Market maker exemption: Market maker positions usually do not participate in ADL sorting to ensure continuous market liquidity.

Malaysian regulators prefer the "profit percentage priority + market maker exemption" combination because it best aligns with the "return-risk matching" principle and prevents market makers from exiting due to ADL risk.

4.3 ADL Execution and Notification

The ADL execution flow must be transparent and auditable:

1. System detects liquidation orders cannot be filled and insurance fund balance is insufficient to cover bankruptcy losses;
2. Calculate total gap amount needed;
3. Reduce positions in ADL priority order from high to low until the gap is filled;
4. Every position reduction must notify affected users in real time, explaining reduced quantity, reduction price, and corresponding loss coverage;
5. All ADL events must be publicly disclosed on the platform announcement page, revealing total reduced amount, number of affected users, and total bankruptcy losses covered;
6. Users can appeal ADL results within 7 days, and platforms must respond within 3 working days.

SC specifically emphasizes that ADL rules must be clearly disclosed before users open positions, and all affected users must be proactively notified after each ADL trigger—no "silent" deleveraging is allowed.

4.4 ADL Caps and Circuit Breakers

To prevent ADL abuse, circuit breaker mechanisms must be in place:

• Single-user daily ADL cap: Maximum 50% position reduction per user per day;
• Platform-wide daily ADL cap: No more than 5% of total platform open interest; if exceeded, suspend trading and trigger circuit breakers;
• ADL interval limit: Same user can be deleveraged at most twice within 24 hours;
• Extreme circuit breaker: If daily ADL exceeds threshold, platform can temporarily reduce market-wide leverage caps to prevent further risk spread.

5. Insurance Fund Design and Management

5.1 Sources of Insurance Fund

The insurance fund is a dedicated capital pool for covering bankruptcy losses. Its sources typically include:

• Platform initial injection: Initial capital injected at platform launch, usually no less than $1 million;
• Liquidation surplus: When liquidation orders execute at prices better than bankruptcy price, 100% of surplus goes to the insurance fund;
• Fee allocation: 5%–10% of every trading fee is automatically injected into the insurance fund;
• ADL surplus: If ADL produces profit beyond the gap, it is injected into the insurance fund;
• Penalty income: Fines from violating users and penalties on market makers.

Malaysian regulators explicitly require: The insurance fund must be completely segregated from platform own funds and client funds, held in an independent custody account, and the platform cannot use it for any other purpose (including operating expenses, market making, investments, etc.).

5.2 Insurance Fund Usage Rules

Insurance fund usage must have clear priorities and approval processes:

1. Liquidation surplus first: Bankruptcy loss from each liquidation is first covered by the surplus from that same liquidation;
2. Contract-level fund second: Each contract can have an independent insurance fund sub-account, prioritizing funds from the corresponding contract;
3. Global insurance fund fallback: If contract-level funds are insufficient, use the global insurance fund;
4. ADL last trigger: Only when the global fund is also insufficient, trigger ADL.

Key rule: The insurance fund cannot be used to cover losses caused by platform bugs or system failures—such losses must be borne by platform shareholders' own capital and cannot use insurance funds formed from user-paid fees.

5.3 Insurance Fund Transparency and Audit

SC requires high transparency for insurance funds:

• Real-time public disclosure of insurance fund balance, historical injection and expenditure records;
• Monthly insurance fund reports disclosing monthly bankruptcy losses, liquidation surplus injections, fee injections, and ADL trigger situations;
• Annual special audit of the insurance fund by an independent audit firm, issuing an audit report;
• When insurance fund balance falls below a safety threshold (e.g., 5x average 30-day bankruptcy losses), the platform must issue a public warning and activate replenishment mechanisms.

SoonTech implemented an "insurance fund transparency dashboard" for clients in Malaysia, allowing anyone to view real-time balances and historical details—this is also a key focus during SC on-site inspections.

6. Bankruptcy Price Calculation and Clawback

6.1 Definition of Bankruptcy Price

Bankruptcy Price is the price at which, if a user's position is closed at that price, the user's margin is exactly zero—the platform neither profits nor loses money. It is one of the most core calculation formulas in the liquidation system.

For a long position, a simplified calculation is: Bankruptcy Price = Entry Price × (1 − Initial Margin Rate). For a 10x long (10% initial margin) entered at 100 USDT, the bankruptcy price is roughly 90 USDT. Once the mark price breaches that level, the position's margin no longer covers its loss, and the remaining shortfall must be absorbed by the insurance fund.

In actual engineering, additional considerations include:

• Cumulative impact of funding rates;
• Realized and unrealized P&L;
• Fee costs;
• Liquidation execution cost (slippage estimation).

If the final liquidation execution price is better than bankruptcy price, surplus funds go to the insurance fund. If worse, the difference is bankruptcy loss borne by the insurance fund or ADL.

6.2 Mark Price and Fair Price

Liquidation cannot be triggered by the latest traded price—otherwise it can be easily manipulated by spoofing and wick attacks. Industry standard is to use "Mark Price" as the trigger basis:

• Mark Price = Median or weighted average of the same contract price across multiple major exchanges;
• Or Mark Price = Spot Index Price + Funding Rate basis (Fair Price);
• Mark Price uses 5–15 minute TWAP to filter instantaneous volatility.

Malaysian regulators explicitly require that Mark Price calculation methods must be public and transparent, data sources must be independent from the platform's own traded prices, preventing platforms from manipulating Mark Prices to maliciously liquidate users.

6.3 Special Clawback Scenarios

Beyond ADL, there are several special bankruptcy clawback scenarios:

• System failure bankruptcy: Bankruptcy losses caused by matching engine, liquidation engine or RPC node failures are borne by platform own capital—insurance fund or ADL cannot be used;
• Price manipulation bankruptcy: Abnormal liquidation caused by malicious Mark Price manipulation requires the platform to roll back trades and compensate user losses;
• Cross-contract chain bankruptcy: Multiple contracts go bankrupt simultaneously during extreme volatility, potentially depleting the insurance fund. In this case, "global bankruptcy clawback" is activated, where all profitable users share losses proportionally, and the platform must inject its own capital to cover at least 20% of the total gap.

7. Liquidation Engine Engineering Architecture

7.1 Three-Tier Architecture: Calculation, Decision, Execution

A production-grade liquidation engine is usually divided into three layers:

1. Calculation Layer: Pulls real-time prices, user positions, margin balances, calculates current margin ratio, liquidation price and bankruptcy price for every position. Typical performance requirement: Full calculation of 100,000 positions < 100ms.
2. Decision Layer: Determines whether to trigger liquidation based on calculation results, which level to trigger, how much quantity to liquidate, and which execution channel to use. Requires state machine support and degradation switches.
3. Execution Layer: Responsible for sending liquidation orders to market maker dark pools, internal matching engines or public markets, monitoring execution status, handling partial fills and cancel-replace flows.

The benefit of decoupled layers: Calculation layer can scale horizontally for performance, decision layer logic changes do not affect the execution layer, and manual takeover is possible if the execution layer has problems.

7.2 High Availability and Fault Tolerance Design

The liquidation engine has one of the highest availability requirements in an exchange—it must work even during extreme volatility, market-wide circuit breakers, and infrastructure failures. Key designs:

• Multi-active deployment: Deploy independently in at least three availability zones—any single zone failure does not affect the whole;
• Local cache fallback: Each engine node caches price and position data locally, can continue working even if the database goes down;
• Degradation switches: Can turn off staged liquidation, turn off dark pool priority, temporarily reduce leverage caps under extreme conditions;
• Manual intervention channel: Risk control team can pause liquidation, adjust parameters, manually take over specific position liquidations from the backend;
• Regular stress testing: Daily full stress tests simulating 10x historical peak liquidation pressure.

7.3 Audit and Observability

Every action of the liquidation engine must leave an immutable audit trail:

• Each liquidation trigger: Record time, mark price, user ID, contract, position size, margin ratio, trigger level;
• Each liquidation execution: Record order ID, sending channel, execution price, filled quantity, slippage, bankruptcy/surplus amount;
• Each ADL trigger: Record gap amount, sorting rule, list of deleveraged users, quantity and price per reduction;
• Each insurance fund change: Record injection/expenditure amount, source/destination, corresponding event ID.

All audit logs must be written to an immutable logging system, retained for at least 7 years, and be accessible to regulators at any time.

8. SoonTech Implementation in Malaysia

8.1 Liquidation Engine Module Overview

SoonTech's delivered liquidation engine for Malaysian compliant exchanges includes:

• Margin calculation service: Supports isolated/cross/hybrid modes, real-time margin ratio and liquidation price calculation for all positions;
• Staged liquidation decision engine: 5-level trigger logic, configurable per-level liquidation ratio and time window;
• Three-tier execution router: Market maker dark pool → internal matching → public market, configurable slippage protection;
• ADL auto-deleveraging module: Configurable sorting rules, circuit breaker thresholds, notifications and auditing;
• Insurance fund management system: Multi-account isolation, automatic injection, transparency dashboard, monthly report generation;
• Regulatory compliance suite: All SC-required reports, audit logs, stress testing tools;
• User notification center: Multi-channel (App Push, email, Telegram, SMS) liquidation warnings and ADL notifications.

8.2 Typical Performance Metrics

In production at a leading Malaysian exchange, SoonTech's liquidation engine actually achieved:

• Single availability zone supports real-time calculation for 500,000+ positions with latency < 50ms;
• Under extreme market conditions (15%+ single-day BTC movement), liquidation success rate 99.7%, bankruptcy rate < 0.05%;
• Average liquidation order slippage < 0.2%, significantly below the industry average of 0.5%–1%;
• ADL trigger frequency: Approximately once every 2–3 months, average user position reduction < 5%;
• Insurance fund adequacy ratio always maintained above 10x (fund balance / 30-day average bankruptcy loss).

8.3 SC Regulatory Alignment

Key SoonTech solution alignments with Malaysian regulatory requirements:

• Retail user protection: Isolated margin by default, 20x leverage cap, staged liquidation, multi-channel warnings;
• Fund segregation: Insurance fund independently custodied, physically segregated from platform own funds;
• Transparency: Public liquidation rules, real-time insurance fund dashboard, public ADL event disclosure;
• Audit requirements: 7-year immutable audit logs, exportable regulatory format reports, stress testing tools;
• Appeal mechanism: Built-in liquidation appeal ticket system supporting user review and platform response.

9. Enterprise Implementation Recommendations

9.1 Pre-Launch Preparation

• Stress testing: Simulate scenarios with 5x historical maximum trading volume and volatility to verify liquidation engine throughput and success rate;
• Market maker agreements: Sign liquidation uptake agreements with at least 2–3 leading market makers, defining uptake obligations and fees under extreme conditions;
• User education: Start user education on liquidation, ADL and insurance funds at least 2 weeks before launch;
• Regulatory communication: Submit liquidation system design documents, parameter configurations and stress test reports to SC for pre-review;
• Disaster drills: Simulate liquidation engine downtime, database failures, price data source anomalies to verify degradation processes.

9.2 Post-Launch Continuous Optimization

• Parameter iteration: Review liquidation parameters (margin rates, stage ratios, slippage thresholds) quarterly based on actual trading data, adjust according to market changes;
• ADL review: Conduct full post-mortem after each ADL trigger, evaluate if sorting rules are reasonable and whether adjustments are needed;
• Insurance fund dynamics: Dynamically adjust fee injection ratio based on historical bankruptcy loss data to ensure fund adequacy always stays within safe range;
• User feedback loop: Collect user feedback on liquidation experience, continuously optimize warning timing, notification channels and liquidation transparency.

9.3 Organization and Process

• Establish a dedicated derivatives risk control team with 7x24 on-call for the liquidation engine;
• Develop emergency plans for extreme market conditions, clarifying who can trigger degradation and who can suspend trading;
• Regularly (at least quarterly) submit derivatives risk reports to SC, including liquidation data, bankruptcy analysis, insurance fund changes and ADL triggers;
• Establish an independent audit position to regularly review liquidation engine code logic against actual execution results.

FAQ

Q1: Does Malaysian regulation have explicit leverage caps for derivatives?

A: The SC RMO DAX framework does not specify exact leverage numbers, but in investor protection guidelines it "recommends" retail user leverage not exceeding 20x and professional investors not exceeding 50x. In practice, if a platform offers 100x leverage to retail users, it will likely be asked to remediate during on-site inspection.

Q2: Does the insurance fund belong to the platform or users?

A: The insurance fund is a common risk reserve for all platform users, not the property of platform shareholders, nor can it be used for dividends, operating expenses or investments. The platform only acts as a trustee managing the fund, must follow preset usage rules, and accepts independent audits.

Q3: Can users refuse to be deleveraged after ADL triggers?

A: No. ADL rules are part of the service agreement users sign when opening positions, and execution is automatic once triggered. But users can appeal—if the platform confirms ADL was triggered due to system bugs or price manipulation, compensation can be provided.

Q4: Why not send all liquidation orders directly to the public market?

A: Liquidation orders are the most "profitable" orders in the market—all bots are front-running and sandwich-attacking them. Sending directly to public markets causes severe slippage, actually increasing bankruptcy risk. Dark pool priority and internal matching are industry standard practices.

Q5: How many data sources should be used for Mark Price?

A: At least 3 independent data sources are recommended, taking median or TWAP. Sources should include at least 2 overseas leading exchanges and 1 local leading exchange, avoiding single data source anomalies causing platform-wide batch liquidations.

Q6: What if even ADL cannot cover bankruptcy losses?

A: This is extreme beyond extreme—historically only FTX-collapse level events cause this. At this point, platforms have two options: shareholders inject their own capital as backstop, or activate "global bankruptcy clawback" where all profitable users share losses proportionally. Malaysian regulators prefer shareholder backstop priority because it protects retail user interests.

Conclusion

The essence of derivatives business is risk management, and the liquidation engine is the last line of defense in risk management. Under Malaysia's RMO DAX framework, the liquidation system is no longer an "it works" edge module—it is the core object of regulatory review, a prerequisite for institutional client onboarding, and critical infrastructure determining platform survival.

A well-designed liquidation engine should complete liquidation silently during extreme market conditions, so that most users do not even feel its existence. A poorly designed liquidation engine can sink the entire platform in a single black swan event.

SoonTech has years of engineering accumulation in CEX derivatives liquidation and Southeast Asia localization, having delivered production-proven liquidation engines for multiple licensed exchange projects in Malaysia, Indonesia and Thailand. If your team is planning or optimizing derivatives business, welcome to communicate with us. We can provide end-to-end solutions from margin models to ADL mechanisms, from insurance fund management to regulatory compliance based on your user structure, market maker depth and regulatory requirements.

Controllable risk, transparent liquidation, trusted platforms—this is the only correct path for long-term derivatives business operations.

🌐 Build secure and scalable Web3 platforms with SoonTech.

Explore our solutions for White Label Crypto Exchanges, Prediction Markets, MPC Wallets, Matching Engines, Liquidity Integration, and Compliance.