OKX is hiring a Principal Engineer, Agent Infrastructure & Memory Architecture in Engineering — APAC. The overview below is synthesized from the employer posting on job-boards.greenhouse.io: factual requirements and scope are preserved, but prose is rewritten with editorial context. Verify details and apply via the employer link.
Who We Are
At OKX, we believe that the future will be reshaped by crypto, and ultimately contribute to every individual's freedom. OKX is a leading crypto exchange, and the developer of OKX Wallet, giving millions access to crypto trading and decentralized crypto applications (dApps).
About the Opportunity
Company context from the listing:
The listing frames the Principal Engineer, Agent Infrastructure & Memory Architecture at OKX as a hands-on product role. This role sits at the intersection of research and production engineering, focusing on scalable long-term memory, knowledge graph infrastructure, and robust context management systems.
You will transform cutting-edge research concepts (e.g., Context Graphs, GraphRAG, agent memory architectures) into auditable, maintainable, and production-grade infrastructure that enables reliable, stateful AI systems.
What You’ll Be Doing
- Architect and implement long-term memory infrastructure for AI agents, including auditable, versioned, and rollback-capable middleware that supports persistent, stateful reasoning.
- Design and own end-to-end GraphRAG and knowledge graph pipelines, from ingestion and schema design to graph indexing, retrieval optimization, and integration with LLM-based reasoning systems.
- Build scalable context management infrastructure, elevating context from prompt-level techniques to structured, maintainable systems such as Context Graphs that enable long-horizon and cross-session reasoning.
- Translate advanced agent research into production-grade systems, prototyping emerging architectures and establishing best practices for memory, retrieval, and infrastructure reliability.
- Ensure reliability, observability, and governance of agent infrastructure, including traceable memory operations, evaluation frameworks, and safeguards against drift and degradation.
What We Look For In You
Experience and skills the team lists as required:
- PhD or Master's in Computer Science, AI, Machine Learning, Robotics or related field with at least 10 years of industry experience.
- Possess a deep expertise in at least one of the following areas: long-term memory, knowledge graphs, or context management.
Nice to Haves
Additional experience noted as a plus:
- Experience in top-tier AI labs or Tech industry leaders
- Experience in high-frequency trading, fintech, Web3, or crypto platforms is a strong advantage
- Startups or research-intensive environments preferred; must be comfortable operating with ambiguity and speed
Perks & Benefits
What the posting highlights about the offer:
- Competitive total compensation package
- L&D programs and Education subsidy for employees' growth and development
- Various team building programs and company events
- Wellness and meal allowances
- Comprehensive healthcare schemes for employees and dependants
- More that we love to tell you along the process!
Please note that Hong Kong is a group-level service hub, and OKX does not carry on a business of operating a virtual asset trading platform in Hong Kong.
Notice:
All official OKX vacancies are published on this website. While roles may appear on selected third-party platforms from time to time, information on other sites may be inaccurate or outdated.
Information collected and processed as part of the recruitment process of any job application you choose to submit is subject to OKX's Candidate Privacy Notice.
Build a standout application
For the Principal Engineer, Agent Infrastructure & Memory Architecture at OKX, reviewers look for concise evidence over buzzwords. Mirror the language of the posting sparingly, quantify support or delivery outcomes, and show how you handled ambiguity, time-zone collaboration and user empathy. Keep your resume to impact, keep your cover note to one page, and link to artifacts — tickets resolved, docs shipped, dashboards owned — that prove you can operate in a fast-moving Web3 team. Prepare to discuss a time you turned a confusing user report into a clear fix and how you measure quality in support and operations.
Web3 hiring values reliability: on-time follow-through, clear writing, and a track record of improving runbooks and tooling. Treat the application as a work sample. For interviews, be ready to walk through how you prioritize across time zones, handle a difficult user, and decide when to escalate versus resolve directly. Show how you document decisions so the next teammate benefits.
In a distributed Web3 org, trust builds through written clarity. Use the cover note to demonstrate it.