<h2><strong>The role</strong></h2> <p>We’re looking for a new teammate to join us on the journey of keeping HelloFresh a trusted name as the business goes through AI transformation across all verticals.</p> <p>The person in this role will lead the security of our AI stack end‑to‑end and make it safe to build boldly. They’ll map real attack paths across data, training, inference, agents, and RAG; turn them into guardrails and launch gates; and enable product teams to ship faster with confidence. Their work protects customer trust, IP, and brand while setting an internal bar for how modern AI should be secured. </p> <h2><strong>What you’ll do</strong></h2> <ul> <li><strong>Threat model and harden</strong> data pipelines, training jobs, inference APIs, agents/tools, and RAG systems</li> <li><strong>Mitigate key risks</strong>: prompt injection, data exfiltration, tool abuse, data/model poisoning, membership inference, model theft</li> <li><strong>Build guardrails</strong>: content safety and policy enforcement, semantic/PII redaction, allow/deny tooling, tenant isolation</li> <li><strong>Operationalize evaluations</strong>: red‑team corpora, adversarial test suites, regression baselines, drift and abuse detection</li> <li><strong>Secure RAG</strong>: document sanitization, namespace/metadata access control, query‑time authZ, cache and retrieval controls</li> <li><strong>Govern secrets and spend</strong> for model providers, embeddings, and vector DBs with quota and anomaly detection</li> <li><strong>Lead IR for AI</strong> features: logging and telemetry design, prompt/session forensics, containment, and post‑incident learning</li> <li><strong>Partner across functions</strong> with Product, ML, Platform, Legal, and Compliance to set launch gates and ship safe features</li> <li><strong>Teach and scale</strong>: internal standards, playbooks, and short courses that raise the baseline</li> </ul> <h2><strong>What you’ll bring</strong></h2> <ul> <li>5+ years in AppSec/Cloud/Platform Security. 2+ years securing ML/LLM apps or building them at scale</li> <li>Proficient in Python and one of TypeScript/Go/Java. Able to read ML code and pipelines</li> <li>Hands-on with modern AI stacks (e.g., OpenAI/Azure AI/Vertex AI/Bedrock) and LLM frameworks (LangChain/LlamaIndex)</li> <li>Strong grasp of cloud IAM, KMS/HSM, container/Kubernetes security, IaC and CI/CD hardening</li> <li>Familiarity with AI security bodies of knowledge: OWASP Top 10 for LLM Apps, MITRE ATLAS, secure RAG patterns</li> <li>Working knowledge of governance and risk frameworks: NIST AI RMF, ISO/IEC 42001 &amp; 23894, GDPR, EU AI Act</li> <li>Clear communicator who drives design decisions and ships secure defaults</li> </ul> <h2><strong>What we offer</strong></h2> <ul> <li>Comprehensive relocation assistance to move to Berlin plus visa application support</li> <li>Healthy discount on weekly HelloFresh boxes</li> <li>Annual learning and development budget to attend conferences or purchase educational resources plus access to the HelloFresh Academy</li> <li>Subsidised childcare with a professional nannying agency</li> <li>Free access to Headspace, biweekly in-house yoga classes (remote since March 2020)</li> <li>A diverse and vibrant international environment of 70+ different nationalities</li> <li>Additional perks include: Free crash course in German, compensation for advanced external German classes, discounts for our neighboring gym &amp; Urban Sports Club, summer &amp; winter parties, discount on our food vending machines</li> <li>The chance to have a significant impact on one of the fastest-growing technology companies in Europe in an exciting growth phase</li> </ul> <p> </p>