Senior Quantitative Developer — High-Frequency Trading Infrastructure (Rust)
About the Company
A Tier-1 global cryptocurrency exchange (Top 10 by volume), serving 30M+ users across 200+ countries. Building out an ultra-low-latency, in-house quantitative trading and market-making platform with a high-performance Rust stack. The HFT team operates at sub-microsecond latencies across market-data ingestion, order execution, and live strategy deployment.
About the Role
We're hiring a Senior Quantitative Developer to own the core trading infrastructure layer and significantly raise the performance and reliability of the HFT platform. This is a hands-on technical leadership role — you'll lead a team of 3–5 engineers and report directly into C-level leadership. The mandate is to build production-grade infrastructure from the ground up (0 1 ownership) and bring institutional-grade execution quality to the in-house trading function.
This is an in-house buy-side role at a major exchange, not a vendor / sell-side / market-making service position. We're looking for someone who has built or operated this kind of system inside a top-tier crypto exchange or proprietary trading firm, and is ready to own the full stack end-to-end.
What You'll Own
- Design, develop, and own high-performance, low-latency components of the HFT stack — market-data handlers, order execution engine, risk gateways — in Rust.
- Implement and optimize low-level primitives: shared memory, lock-free / wait-free data structures, zero-copy networking, custom TCP/UDP protocols, and Linux kernel / user-space tuning.
- Drive end-to-end performance: latency profiling, throughput optimization, and production stability of the entire trading pipeline.
- Build and maintain high-performance backtesting and simulation frameworks supporting quantitative researchers.
- Lead and mentor a team of 3–5 engineers; own technical direction, code quality standards, and architectural decisions.
- Collaborate closely with quantitative researchers, infrastructure teams, and senior leadership to translate strategy requirements into reliable production systems.
- Own system observability and participate in on-call rotation for live trading infrastructure.
What We're Looking For — Must Have
- 5+ years of professional Rust development. Exceptionally strong C++ background combined with proven Rust proficiency is also acceptable.
- 2+ years of direct experience in a high-frequency trading (HFT) or ultra-low-latency production environment.
- Direct experience building trading infrastructure from 0 1 — not just maintaining or extending existing systems.
- Deep low-level systems expertise: Linux internals and networking stack (epoll, io_uring, TCP/UDP tuning, kernel bypass), shared memory, lock-free / wait-free data structures, zero-copy techniques, high-performance IPC.
- Proficiency in multithreading, concurrency models, cache-friendly design, and performance profiling tools (perf, flamegraphs).
- Experience designing or maintaining core infrastructure components: custom message buses, high-speed caches, persistent logging.
- Demonstrated ability to lead and mentor a small engineering team.
- Bachelor's degree or higher in Computer Science, Engineering, or related quantitative field (or equivalent demonstrated experience).
Strongly Preferred
- Prior experience at a Tier-1 crypto exchange (Binance, OKX, or equivalent buy-side / in-house quant team).
- Experience using AI agents in daily development workflows — agentic harnesses, AI coding assistants, or autonomous dev tools to accelerate Rust development, testing, and refactoring.
- Familiarity with agentic development practices or frameworks supporting complex, multi-step development tasks.
- Hands-on experience with financial market data protocols (FIX, WebSocket) and low-latency order execution pipelines.
- Cloud deployment experience (AWS or equivalent) for non-latency-critical components: backtesting clusters, monitoring, observability stack.
- Modern observability tooling in high-throughput environments (Prometheus, Grafana, OpenTelemetry).
Compensation & Logistics
Compensation: Approximately RMB 2M / year total (negotiable based on seniority and equity preference). Competitive package including base + performance variable.
Location: Shenzhen (on-site) or Hong Kong (on-site / hybrid). Hong Kong requires existing work authorization — no visa sponsorship for HK.
Work Culture: Performance-driven environment, but not a 996 culture. Standard professional working hours with on-call rotation for production systems.
Team: Lead 3–5 engineers. Report directly into C-level leadership.
某全球Top 10加密货币交易所（按交易量排名），用户规模超3000万，覆盖200+国家和地区。目前正在搭建自研超低延迟量化交易及做市平台，核心技术栈基于Rust，HFT团队在市场数据接收、订单执行、实时策略部署等环节实现亚微秒级延迟。
我们正在招聘一位 高级量化开发工程师 ，负责核心交易基础设施层的搭建与所有权，全面提升HFT平台的性能与稳定性。这是一个hands-on的技术领导岗位 —— 你将带领3-5人工程团队，直接汇报至C-Level领导层。岗位定位为 0 1搭建 生产级交易基础设施，将机构级别的执行质量引入自营交易体系。
这是 甲方in-house买方角色 ，不是vendor / sell-side / 做市服务类岗位。我们寻找的是在Tier-1加密交易所或顶级自营交易团队中真正搭建或运营过此类系统的人才，能够独立owner整个技术栈。
- 设计、开发并负责HFT核心低延迟组件（市场数据处理、订单执行引擎、风险网关等），技术栈为Rust
- 实现并优化底层基础组件：共享内存、无锁/无等待数据结构、零拷贝网络、自定义TCP/UDP协议、Linux内核及用户态调优
- 推动端到端性能优化：延迟剖析、吞吐量提升、生产环境系统稳定性
- 开发并维护高性能回测与仿真框架，为量化研究团队提供支持
- 带领并mentor 3-5人工程团队，负责技术方向、代码质量标准、架构决策
- 与量化研究员、基础设施团队、高管层紧密协作，将策略需求转化为可靠的生产级系统
- 建设系统可观测性，参与生产交易系统On-Call值班
- 5年以上Rust实际开发经验 （如C++背景极强且Rust熟练度高也可接受）
- 2年以上高频交易（HFT）或超低延迟生产系统经验
- 具备0 1搭建交易基础设施经验 —— 不是维护或扩展已有系统
- 扎实的低层系统功底：
- 深入理解Linux内核与网络栈（epoll、io_uring、TCP/UDP调优、内核旁路等）
- 熟练使用共享内存、无锁/无等待数据结构、零拷贝技术及高性能IPC
- 精通多线程、并发模型、缓存友好设计、性能分析工具（perf、flamegraph等）
- 有设计或维护核心基础设施组件的经验（自定义消息总线、高速缓存、持久化日志等）
- 有带3人及以上工程团队的经验
- 本科及以上学历，计算机、软件工程或相关理工科专业（或具备同等能力）
- 有Tier-1加密交易所（Binance、OKX或同等水平的甲方in-house量化团队）工作背景
- 在日常开发工作中使用AI Agents的经验（agentic harness、AI编码助手、自动化开发工具等，用于提升Rust项目的代码生成、测试、重构效率）
- 熟悉支持复杂多步开发任务的agentic开发实践或框架
- 熟悉金融市场数据协议（FIX、WebSocket等）及低延迟订单执行流程
- 云平台部署经验（AWS或同类平台），负责非核心延迟组件（如回测集群、监控、观测性系统）
- 熟悉高吞吐环境下的现代可观测性工具（Prometheus、Grafana、OpenTelemetry等）
- 薪资 ：约 人民币200万/年 全包（可议，根据资历及股权偏好调整），具备竞争力的base + 绩效variable组合
- 办公地点 ：深圳（on-site）或香港（on-site / 混合办公）； 香港岗位需候选人已具备本地工作签证，公司不提供香港签证赞助 ；深圳可作为备选
- 工作文化 ：高绩效环境， 非996文化 ，标准职业工作时间，生产系统轮值On-Call
- 团队 ：带领3-5人工程团队，直接汇报至C-Level领导层
