About

Joaquin Bejar

Quantitative engineer with more than 20 years building low-latency trading platforms for electronic markets and digital assets. I work mostly in Rust, together with C++, Go and Python.

My work covers the whole trading core: order books and matching engines, options pricing and real-time risk, smart order routing and transaction-cost analysis, market making and exchange connectivity over FIX, SBE, REST and WebSocket. Much of it is published as open source. I am based in London.

Live demo

Chainwalk

Watch an options position breathe.

A simulator that walks a multi-leg options position through a stochastic market, step by step, showing the P&L and twelve Greeks as they move. Every price is a bid or an ask, so spread and commission are charged on entry, and every run is reproducible and shareable.

  • Many expirations and strikes alive at once on one market tape
  • Ten stochastic models, from geometric Brownian motion to Heston and GARCH
  • A study API for Monte Carlo reports over thousands of paths

Rust · TypeScript · OptionChain-Simulator · Monte Carlo

Open Chainwalk

What I work on

Selected work

Systems I have built for trading firms and exchanges. Client and employer names are left out.

  • 2025–2026

    Hybrid exchange for options structures

    A non-custodial exchange for crypto options and multi-leg structures: order matching, risk, liquidation and RFQ run off-chain in Rust, while custody, clearing and settlement happen on-chain. I led the engineering team that built it and wrote large parts myself: the matching engine, the options-chain order book, the API gateway and the protocol and market-data layers.

    Matching engine
    One sequencer per underlying, price-time priority and position netting, persisted in PostgreSQL with hot state in memory.
    Options-chain order book
    Lock-free underlying → expiry → strike → contract hierarchy on top of OrderBook-rs, PriceLevel and OptionStratLib.
    Risk engine
    Scenario-based portfolio margin in decimal arithmetic, synchronous pre-trade checks over SBE, account health states and what-if simulation.
    Liquidation engine
    Dutch-auction liquidations, backstop liquidity allocation, an insurance fund and auto-deleveraging, with signed results settled on-chain.
    Account management
    Accounts, collateral and cross-collateral balances shared by the trading and risk services.
    API gateway
    Single entry point for every service: signed orders and token authentication verified at the edge, identity propagated downstream.
    OTC RFQ engine
    Request-for-quote auctions that combine market-maker quotes with on-chain liquidity for block trades and complex structures.
    Protocols and market data
    SBE binary streaming for market makers and order entry, REST for control and accounts, WebSocket subscriptions, and a market-data fan-out with snapshots and resync.

    Rust · SBE over TCP · REST · WebSocket · NATS JetStream · PostgreSQL · Redis · Solidity

  • 2026

    Pre-trade risk engine

    Checks every order before it leaves: capital reservations backed by a write-ahead log for crash recovery, exposure tracking, policies reloaded without restarts and a kill switch, within a sub-millisecond budget.

    Rust · write-ahead log · ClickHouse

  • 2026

    Multi-exchange execution service

    Smart order router and order management system across nine centralized exchanges (spot, perpetuals and options): order lifecycle, fill de-duplication, reconciliation and circuit breakers, with interchangeable transports behind one event bus.

    Rust · gRPC · REST · NATS

  • 2026

    Strategy engine

    A Rust host that runs Python trading strategies in-process, with a local risk layer that can only tighten global limits and order routing to several execution services.

    Rust · Python · PyO3

  • 2025–2026

    Market-data pipeline

    Order-book feed handlers for about ten trading venues, centralized and on-chain, plus ingestion of decentralized-exchange pool events, with configuration changes applied without restarts. Co-author; led several exchange integrations.

    Rust · ClickHouse

  • 2026

    Liquidity analytics API

    Metrics behind a public liquidity rating: price gap between centralized and decentralized venues, order-book resilience, fillability and cost to trade, served with caching and live streams.

    Rust · ClickHouse · Redis

  • 2026

    Equities order management with algorithmic execution

    TWAP and VWAP slicing with adaptive aggressiveness, broker-side stops, event sourcing and a risk actor that can veto every child order, plus a market-data service with vendor failover. Sole author.

    Rust · NATS JetStream · gRPC

  • 2025–2026

    Price-triggered order management

    Trigger orders with a time to live that fire when best bid or offer crosses a level, then generate linked take-profit and stop-loss orders and hand the parent to execution.

    Rust · NATS · Redis

  • 2025

    Prediction-market order book

    Matching service for YES/NO outcome markets: one book per market, a controlled market lifecycle and snapshot-based recovery, built on OrderBook-rs.

    Rust · PostgreSQL · OrderBook-rs

Publications

  1. 2026 · Industry whitepaper

    Structured Liquidity: An OTC Framework for Event-Driven Crypto Markets

    Joaquin Bejar Garcia. SSRN.

    DOI 10.2139/ssrn.5920562 · SSRN · ResearchGate

  2. 2018 · Technical report

    INFORM scientific and technical improvements in 2017: Missing values imputation and IT developments

    Montserrat Marin Ferrer, Brian Doherty, Joaquin Bejar Garcia, Stefano Luoni, Luca Vernaccini. Publications Office of the European Union.

    DOI 10.2760/076136 · DOI · JRC repository

Selected open source

All 67 repositories

OrderBook-rs (GitHub)

Thread-safe limit order book and matching engine in Rust, designed for low-latency concurrent access.

Rust · 539 stars · order books

OptionStratLib (GitHub)

Options pricing and strategy library: Black-Scholes, Greeks, strategy payoffs and analysis.

Rust · 251 stars · options

market-maker-rs (GitHub)

Quantitative market making strategies, starting with the Avellaneda-Stoikov model.

Rust · 102 stars · market making

IronSBE (GitHub)

Simple Binary Encoding (SBE) codec with server and client, for low-latency messaging.

Rust · 21 stars · protocols