Overview
We are building a confidential computing AI platform from the ground up, a containerized, GPU-accelerated infrastructure designed to securely run large language models, multi-agent systems, and advanced AI workloads in zero-trust environments. As AI Security Engineer, you will be part of the founding team responsible for designing and implementing the platform's end-to-end security architecture, from confidential computing and enclave orchestration to secure agent execution, attestation flows, and runtime integrity. This is not a maintenance role; this is an opportunity to build secure AI from first principles. Your focus will be on data in use, with safeguarding for data in transit and data at rest, ensuring robust encryption, access control, and integrity across the entire AI data lifecycle. You will collaborate with AI engineers, infrastructure teams, and product leadership to translate our vision into a real, working, secure platform that can scale from pilot deployments to global, high-stakes use cases.
Responsibilities
Help design and implement the secure foundation of our AI platform, from enclave orchestration to attestation and encrypted data pipelines.
Build out and harden our Confidential Containers stack (Kata, CoCo) for secure workload execution inside Kubernetes clusters.
Develop and maintain remote attestation pipelines and verifiable compute infrastructure for agentic and LLM workloads.
Secure the interaction between users, models, and data through encrypted connectors, secure API gateways, and TLS-protected ingress layers.
Implement comprehensive protections across data at rest, data in transit, and data in use, ensuring confidentiality and integrity from storage to inference.
Identify and mitigate potential threats specific to LLMs and multi-agent systems, such as prompt injection, task hijacking, and data leakage.
Collaborate on system-wide trust and integrity models, ensuring isolation across workloads, containers, and GPUs.
Contribute to the definition of platform-level security best practices and audit-ready documentation.
Support internal teams and pilot partners with secure deployments, validations, and future accreditations.
Qualifications
Bachelor's or Master's degree in Computer Science, Cybersecurity, Engineering, or a related technical field.
5+ years of experience in security engineering, preferably with cloud-native systems, AI infrastructure, or confidential computing.
Familiarity with Trusted Execution Environments (TEEs) such as Intel TDX, AMD SEV-SNP, ARM CCA.
Practical experience with Kubernetes security, container orchestration, and hardened Linux environments.
Understanding of cryptographic security protocols and systems (TLS, attestation, encrypted storage).
Hands-on experience or deep familiarity with securing data at rest and in transit, alongside data-in-use protections through enclave technologies.
Experience with CPU-GPU enclaves (or interest in researching H100, SEV-GPU, etc.).
Knowledge of AI safety practices, agent isolation, or privacy-preserving machine learning is a bonus.
A hands-on, systems-level builder with strong debugging skills and a proactive, team-oriented mindset.
Comfort operating in fast-paced, high-autonomy environments where ownership and initiative are key.
What We Offer
The opportunity to build a secure AI platform from scratch, shaping foundational infrastructure for Europe's digital sovereignty.
A high-impact role working at the intersection of AI, hardware, and deep infrastructure security.
Direct involvement in one of Europe's most ambitious confidential computing and AI safety efforts.
A chance to contribute to open-source infrastructure, with dual-use applicability across sectors.
A flexible, hybrid work environment, with opportunities for growth, mentorship, and long-term leadership.
At Openchip & Software Technologies S.L., we embrace diversity and inclusion. We foster a work environment where everyone feels valued, respected, and empowered to reach their full potential – regardless of race, gender, ethnicity, sexual orientation, or gender identity.
#J-18808-Ljbffr