Role Overview

Consult and review secure architectures for our AI systems – from in-house models to third‑party LLMs (incl. RAG, vector databases, APIs, and integrations into our products and internal tools).
Conduct AI-specific threat modeling and security reviews across the ML lifecycle (data → training → deployment → monitoring).
Perform security testing / red-teaming of LLM and ML systems (e.g. prompt injection tests, jailbreaks, exfiltration and data-leakage tests).
Work closely with data scientists, Machine Learning engineers, platform engineers and Compliance & IT Security to define and implement concrete controls in pipelines, infrastructure and applications.
Own and support AI risk assessments, and help write/review policies, standards and governance documentation for AI use.
Translate EU AI Act, financial-services regulation and relevant standards into practical technical and process controls.
Help define monitoring, logging and incident response for AI/LLM systems, including misuse and data-leak detection.
Collaborate with Legal, Compliance and Procurement on AI vendor selection, risk assessments and contract reviews.

Requirements

Demonstrable experience in Artificial Intelligence/Machine Learning security in a production context – not just general cybersecurity.
Practical knowledge of LLM-specific risks, such as: prompt injection and jailbreaks data leakage and sensitive information exposure model inversion, membership inference supply chain risks in AI tooling and models.
Solid understanding of the ML lifecycle and typical MLOps setups (data pipelines, training, evaluation, deployment, CI/CD, monitoring) and where to place security controls.
Experience designing or reviewing secure architectures for AI/LLM systems, including: API security and authentication/authorization secrets management (API keys, tokens, credentials) isolation of tenants/contexts and access control for data sources & vector stores protection of sensitive data in prompts, logs and training data.
Experience working side-by-side with data scientists or ML engineers – you have credibility in technical rooms and can challenge design decisions constructively.
Ability to read Python code and basic ML pipelines and to build small scripts/tools (e.g. for automated tests, log analysis, or prototype guardrails).
Background in risk assessment and in writing or reviewing policy and governance documentation.
Familiarity with relevant AI standards and frameworks, such as: ISO 42001 OWASP LLM Top 10 NIST AI RMF OECD AI Principles.
Understanding of EU AI Act obligations and how they apply to a fintech / financial services context, with the ability to map them to concrete controls.
Strong grasp of data protection and privacy-by-design in AI (data minimisation, pseudonymisation/anonymisation, retention and deletion of training and log data)

Tech Stack

Cyber Security
Python

Benefits

Strong experienced international team to support and mentor you along the way, smooth onboarding process
International team of 30+ nationalities with professionals and experts
Flat hierarchy, transparent and appreciative feedback culture, monthly all hands meetings, annual feedback and evaluation cycle, regular 1-on-1s with your lead
Well-structured onboarding process as well as supportive and welcoming colleagues
Personal learning & development budget as well as German and English language courses
Good salary for your strong performance
Unlimited employment contract, flexible working hours and 28 vacation days for your work-life balance
Company pension plan, partly covered Deutschlandticket (public transport) and access to “Corporate Benefits” voucher platform to ensure your full well-being
Fun company summer and Christmas parties as well as regular team events

PAIR Finance

Open Positions (1)

Engineering

Senior AI Security Engineer