Senior AI Security Engineer
Firmus Technologies1 month ago
Sydney, AustraliaSenior
Responsibilities
- Define secure-by-design architectures for AI-powered products and services from discovery through production.
- Threat-model LLM-enabled and agentic workflows, including prompt injection, data exfiltration, unsafe tool use, excessive permissions, cross-tenant access, and unintended autonomous actions.
- Design controls for agent identity, authorization, credentials, tool allowlists, approval gates, execution sandboxing, rollback, auditability, RAG, enterprise data, and tenant isolation.
- Secure user-facing AI products and model-serving APIs through authentication, authorization, quotas, validation, monitoring, logging, abuse detection, and data-protection controls.
- Build reusable AI security guardrails, libraries, reference architectures, templates, policy-as-code, and developer tooling.
- Secure Kubernetes, job-scheduler integrations, CI/CD pipelines, secrets management, containers, runtime environments, and GPU software components.
- Lead vulnerability management for AI application dependencies, agent frameworks, APIs, containers, model-serving runtimes, CUDA, and related infrastructure.
- Embed security requirements into design reviews, pull requests, CI/CD controls, releases, runbooks, and incident-response procedures.
- Develop detections and telemetry for anomalous agent actions, unsafe tool calls, unusual data retrieval, credential misuse, API abuse, and policy violations.
- Act as the principal security liaison between AI & Applications and Cybersecurity, coordinating reviews, exceptions, compliance evidence, incident response, and remediation tracking.
Requirements
- At least 5 years of experience in security engineering, application security, cloud security, platform security, or a related field.
- Experience embedding security practices into software engineering or product-development teams and supporting secure delivery through production.
- Strong knowledge of application security, API security, secure software development lifecycle practices, threat modeling, vulnerability management, and security automation.
- Practical knowledge of AI and LLM application security risks, including prompt injection, jailbreaks, insecure tool use, excessive agency, model abuse, sensitive-data exposure, cross-tenant leakage, and supply-chain risks.
- Experience securing agentic applications, RAG systems, enterprise search, vector databases, model-serving APIs, enterprise-data connectors, or comparable AI-enabled systems.
- Experience designing identity and access controls using OIDC, OAuth 2.0, SSO, RBAC, ABAC, workload identity, service accounts, secrets management, and least-privilege principles.
- Strong Kubernetes and container-security knowledge, including admission controls, network policies, RBAC, runtime protection, image scanning, software supply-chain security, and policy engines.
- Experience with secure CI/CD, dependency governance, SBOMs, image signing, provenance, artifact management, and controlled-release processes.
- Proficiency in Python, Go, or a similar language for automation, security tooling, integrations, and analysis.
- Familiarity with cloud and infrastructure security, logging, monitoring, incident response, encryption, key management, and multi-tenant security controls.
- Knowledge of NIST, ISO 27001, SOC 2, CIS Controls, OWASP ASVS, OWASP API Security Top 10, or equivalent frameworks.
- Strong cross-functional communication, ownership, judgment, and ability to work in a fast-moving AI product environment.
Benefits
- Work location is Singapore or Australia.
- Work alongside founders and experts in AI infrastructure, energy systems, and next-generation compute.
- Opportunity to gain experience as part of an NVIDIA Cloud and Engineering partner in Asia Pacific.
- Founder-led environment with accessible leadership, fast decisions, low bureaucracy, and early ownership.
- Opportunity to contribute to sustainable AI infrastructure and energy-efficient AI Factories.
Tech Stack
Categories
About Firmus Technologies
Firmus Technologies builds energy‑efficient AI infrastructure, developing liquid‑cooled “AI Factory” data centers and operating a large‑scale GPU cloud for model training. The company sells capacity and services to developers, enterprises, education, and government customers, with a focus on energy and cost efficiency across Asia‑Pacific. Founded in 2019 in Australia, Firmus is privately held and headquartered in St Leonards, Tasmania.