Senior Security Engineering

Location: Zinacantepec

Join Ciberia Tech and help redefine the future of AI-driven cybersecurity

At Iris we do more than build companies — we shape leaders in technological innovation. Our mission is to reshape our clients' future through the most advanced technology available:
Cloud, Cybersecurity, Data and AI. What drives us is ambition, excellence, innovation, transparency, empathy and trust.

We invite you to become part of Ciberia Tech, the new global reference in managed cybersecurity services, powered by Artificial Intelligence and backed by Google Cloud Security.

We are a Google Premier Partner for Cybersecurity across EMEA and LATAM, operating in five countries. If you are looking for a place where your professional judgement shapes strategic engagements and where you work with genuinely leading-edge technology, you have found it.

We are seeking a Senior Security Engineering professional to join our AI Security practice and take end-to-end ownership of the AI/ML lifecycle—from training data integrity to production endpoint security. In this role, you will harden AI platforms, safeguard training corpora, and engineer defenses against adversarial and generative AI threats. Beyond the technical execution, you will serve as a trusted advisor to our clients, helping them assess real-world exposures and prioritize strategic remedies.

What you will be doing:

- Define and deploy the control set that protects AI and ML pipelines end to end, so that models and datasets retain their integrity and confidentiality from ingestion through to production serving.
- Lead threat modelling exercises and technical assessments aimed specifically at AI estates, paying particular attention to model exfiltration, training-data poisoning and prompt-injection vectors.
- Work shoulder to shoulder with data science and platform engineering teams to turn adversarial-machine-learning countermeasures into day-to-day operational practice, and to apply privacy-preserving techniques such as differential privacy where they belong.
- Build the monitoring and detection layer for AI-specific telemetry in Security Command Center Enterprise, and wire those signals into our managed service so that what reaches the client is qualified insight rather than volume.
- Design and run enablement sessions on secure AI development for client engineering teams, and raise the technical bar inside our own delivery squads while you are at it.
- Produce reusable assets — reference architectures, assessment frameworks, response playbooks — that hold up when applied from a single engagement to a multi-client managed service.
- Act as the escalation point on AI risk for pre-sales conversations and strategic client discussions, including at executive level.

What do we expect from you:

- A solid foundation in security engineering, with demonstrable specialisation in AI and ML security — model protection and adversarial machine learning in particular.
- A track record of securing AI platforms and cloud-native services on Google Cloud, including hands-on work with Vertex AI, GKE and the services around them.
- Command of data protection obligations and, more to the point, of how to implement them technically over training datasets at scale.
- The ability to lead a hands-on technical workshop, and to translate AI risk into language an engineering team and an executive committee can each act on.
- Comfort working with incomplete information: much of this field has no settled playbook yet, and clients will expect you to form a defensible position anyway.

What would set you apart:

- Google Cloud Professional Machine Learning Engineer or Professional Cloud Security Engineer certification.
- Prior work with recognised AI security frameworks, and experience running AI-focused red team exercises or audits.
- Experience taking generative AI and large language model workloads safely into production, including guardrail and evaluation design.
- Exposure to a managed service or consulting delivery model with several concurrent clients.
- Contribution to the wider community — research, published tooling, conference talks or open-source work on AI security.