About the Role
We’re looking for a Principal SW Security Software Engineer to secure large-scale AI inference infrastructure running complex, high-performance workloads. You’ll design and implement security controls across the entire inference stack — from APIs and control planes to distributed systems, accelerators, and runtime environments. This role blends systems engineering, platform security, and operational rigor in an environment where performance and reliability are mission critical. This is a hands-on engineering role with real ownership over production systems.
What You’ll Do
- Architecting, implementing, and scaling robust security solutions for AI inference platforms
- Lead and drive strategic security initiatives, fostering secure by default developer experiences, and mentoring high performing engineering teams.
- Secure model serving, APIs, and control planes against abuse and attacks
- Build authentication, authorization, and identity systems for internal and external services
- Implement network security (service isolation, mTLS, zero-trust patterns)
- Harden inference runtimes, containers, and host systems
- Develop protections against:
- Build secure key and secret management systems
- Partner with infra and ML teams to bake security into system design
- Perform threat modeling and security reviews for new features
- Lead incident response for security events related to inference systems
Core Engineering Qualifications
- 10+ years of strong software engineering background (not just policy or compliance)
- Proficiency in one or more of:
- Go, Rust, C++, Java, Python
Security Experience
- Experience securing production infrastructure or platforms
- Authentication & authorization (OAuth, OIDC, RBAC, ABAC)
- TLS, mTLS, cert management
- Secure service-to-service communication
- Hands-on experience with:
- Secure boot / host hardening
- Supply chain security (SBOMs, signing)
AI / Inference-Specific (Highly Valued)
- Securing model serving pipelines
- Protecting inference APIs from:
- Multi-tenant inference environments
- Hardware accelerators (GPUs, custom silicon)
- Performance vs security tradeoffs
- Experience working with ML or AI platform teams
