Senior AI Full-Stack Applications Engineer, OKX

• 6+ years building and maintaining full-stack web applications in production, owning them over time
• Strong engineering fundamentals: data structures and algorithms, fluent Python (or a comparable backend language), and strong SQL and data modeling. True full-stack: from the data layer and SQL through resilient APIs to the UI, wired into the cloud (AWS or GCP)
• Strong front-end (React / Next.js and TypeScript, with real design and UX sense) and proven real-time / streaming data visualization with drill-through to underlying evidence
• Agentic AI features with responsible-AI practice: multi-step workflows, human-in-the-loop review, output validation, red-teaming, and honest handling of hallucination and error
• Data protection at the app layer and audit-trail thinking (access controls, evidence, defensible handling), backed by automated testing and documentation. You build applications that could withstand external audit, by design. You ship inside locked-down enterprise environments (TLS-intercepting proxies, endpoint detection and response tooling, restricted installs, security guardrails) without treating security as someone else’s problem
• Experience integrating third-party systems (JIRA, Google Workspace, SaaS APIs) and partnering with non-technical users to drive adoption

Desirable

• OKX’s Internal Audit function has an early-stage, working AI-native capability: live dashboards, agentic workflows, and AI-powered apps that let a very small team punch far above its weight
• Your job is not to rebuild it at today’s maturity
• Your job is to take it to a regulator-grade production standard and well beyond, and to drive AI Enablement across the department
• You own the products people touch: the real-time dashboards that drive Chief Audit Executive and board reporting, the computer-assisted audit technique (CAAT) and continuous-monitoring surfaces auditors work from, and the agentic, AI-powered apps that make the team faster, and you get auditors to adopt them
• You build on the platform’s runtime and harness rather than reinventing them. We hire on demonstrated building, not on claims or credentials
• We expect you to be more capable than the hiring manager in your domain: you will co-own and challenge the tooling strategy, not just implement it
• This is a two-person engineering team: you deploy, debug, and hotfix your counterpart’s stack when needed
• In your first year:
• A live, real-time Chief Audit Executive and board dashboard is in use for actual reporting
• CAAT and continuous-monitoring surfaces are used by auditors in multiple real audits, with full-population testing live in at least two audit domains
• Agentic, human-in-the-loop applications are part of the team's standard workflow: several in production, adoption measured and growing, and at least two demonstrably cutting cycle times, one in audit reporting and one in a board-cycle deliverable
• The hiring manager is out of the delivery loop for dashboards, reports, and rendering pipelines: each has a runbook and an owner who is not him
• How we assess demonstrated building
• We assess demonstrated building in ways that respect your time and your confidentiality obligations to current and former employers: a portfolio deep-dive (walk us through applications you built and kept running, at the level of detail your obligations allow; we want architecture, decisions, and trade-offs, never proprietary code, data, or documents), a short, time-capped build exercise on a synthetic dataset unrelated to OKX's business (a live dashboard with drill-through, used for assessment only, never put to use by OKX, and the work remains yours) that you defend live, walking us through your design decisions and extending it on the spot, a working session on how you drove adoption with non-technical users, and references focused on whether you built and maintained applications in production
• Trust and compliance:
• This role handles highly sensitive audit data at a global crypto exchange
• Expect background checks, confidentiality obligations, and personal-trading and material-non-public-information restrictions consistent with OKX policy
• Environment:
• Cloud on AWS or GCP. Claude as the starting point in a deliberately multi-model architecture: Anthropic, OpenAI, Google, using the right model for each job, including plugins you build
• Google Workspace for reporting and evidence
• JIRA for the audit team’s workflow. Lark for team communications and alert bots
• Today’s executive product ships largely as API-rendered Google Docs/Slides and branded PDF; you evolve that toward real applications without dropping the document standard. React / Next.js front-end, with testing and observability throughout
• Treat this stack as the starting point, not a constraint: you build production-grade products end to end on what exists today, and you are expected to propose, prove, and adopt better tools as demand and capabilities evolve