Location
Based in the Limassol office in Cyprus, with full relocation support available for you and your family.
About the company
A global fintech and trading broker operating at scale across many countries, serving over a million clients. The team focuses on combining technology, care, and long-term thinking to improve trading for clients worldwide.
Responsibilities
- Operate and secure a centralized AI gateway as the single controlled path to SaaS and on-prem models, including routing by data classification and security telemetry.
- Build guardrails, policy enforcement, and fine-tuned classifiers against prompt injection and data leakage, and run least-privilege security review and onboarding of MCP servers.
- Design isolation and access controls so AI agents act only within their users' permissions, with risk-scored human-in-the-loop approval for critical actions.
- Detect and bring Shadow AI under control, and automate security review, MCP and AI tool assessment, and AI security event analysis using LLMs and agents.
Requirements
- Bachelor's or Master's in Computer Science, Software Engineering, Information Security, Applied Mathematics/ML, or equivalent practical experience.
- 5+ years in security engineering, platform/infrastructure engineering, or backend development with a strong security focus, plus 2+ years of hands-on production work with LLM/GenAI systems.
- Hands-on LLM engineering: fine-tuning open-weight models with PyTorch and Hugging Face (Transformers, PEFT, TRL), serving and optimizing inference with vLLM on Kubernetes, and building evaluated LLM-powered automations or agents.
- Strong Python or a systems language (Go, Rust, C/C++) for production services, with Linux/Kubernetes security, identity (OAuth 2.1/OIDC, JWT), policy-as-code (OPA/Rego), IaC, and knowledge of AI-specific attacks, MCP, OWASP LLM/Agentic Top 10, and MITRE ATLAS.