Software Engineer – Platform Security
FriendliAI
About the role
About the job
FriendliAI is looking for an engineer to architect and build the security foundations of a multi-tenant serverless compute platform running on Kubernetes with hardened container isolation. As a Software Engineer, Platform Security, you will own the security design for both the control plane and data plane, implement guardrails as code, and partner closely with Platform/SRE/Infra teams to ship a secure-by-default developer experience. This is a hands-on builder and architect role.
Key Responsibilities
Design and implement the security architecture for a Kubernetes-based, multi-tenant serverless platform.
Build guardrails as code using Terraform and Helm.
Establish network segmentation and service-mesh policy.
Develop secrets and key management patterns.
Implement runtime detections for containerized workloads.
Design and roll out IAM & workload identity management.
Threat model new features and changes to the overall platform.
Collaborate with engineering to debug production issues, lead post-incident hardening, and automate evidence for compliance controls.
Qualifications
5+ years securing cloud-native platforms with a focus on distributed systems and multi-tenancy.
Bachelor’s or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent.
Strong hands-on experience with AWS (IAM, VPC/networking, logging/monitoring, …) and Kubernetes/container security.
Proficiency with Terraform and Helm.
Experience building or operating runtime detections for containers/functions.
Strong technical background in backend systems, cloud infrastructure, or AI tooling
Clear written and verbal communication; ability to translate pricing/business rules into robust systems.
Preferred Experience
Knative or similar serverless-on-K8s frameworks.
Container hardening and sandboxing.
Policy-as-code and admission policy design at scale.
eBPF-based observability, detections, and signal tuning.
Multi-cloud exposure and cross-cloud identity approaches.
Hands-on contributions to SOC 2 / HIPAA control automation and audit evidence pipelines.
Benefits
A front-row seat to the generative AI infrastructure revolution
Competitive compensation and benefits package
Daily lunch and dinner provided; unlimited snacks and beverages
Health check-up and top-tier hardware support
Flexible working hours and a highly collaborative environment
About us
FriendliAI is building the next-generation AI inference platform that accelerates the deployment of large language and multimodal models with unmatched performance and efficiency. Our infrastructure powers high-throughput, low-latency workloads for global organizations and integrates directly with Hugging Face, providing instant access to over 600,000 open-source models. We are on a mission to deliver the world’s best platform for AI inference.
Responsibilities
- Design security architecture for a Kubernetes-based, multi-tenant serverless platform.
- Build security guardrails as code using Terraform and Helm.
- Establish network segmentation, service-mesh policies, secrets management, and workload identity.
- Implement runtime detections for containerized workloads.
- Threat model platform features and changes.
- Partner with engineering teams to resolve production issues, harden systems after incidents, and automate compliance evidence.
Qualifications
- 5+ years securing cloud-native platforms, especially distributed and multi-tenant systems.
- Bachelor’s or master’s degree in a relevant engineering or computer science field, or equivalent experience.
- Hands-on experience with AWS, Kubernetes, and container security.
- Proficiency with Terraform and Helm.
- Experience building or operating runtime detections for containers or functions.
- Strong background in backend systems, cloud infrastructure, or AI tooling.
- Clear written and verbal communication skills.
Benefits
- Competitive compensation and benefits package.
- Daily lunch and dinner, with unlimited snacks and beverages.
- Health check-up and hardware support.
- Flexible working hours and a collaborative environment.
- Opportunity to work on generative AI infrastructure.
Skills mentioned
About FriendliAI
FriendliAI is the Frontier Inference Cloud for Agents, delivering high throughput, low latency, and reliability at scale for agentic workloads. Through vertically optimized inference infrastructure, it achieves 2–5× faster output token speed and 50–90% lower inference cost, backed by a 99.99% uptime SLA built for high-volume agent traffic. The platform powers low-latency streaming for real-time agents, reliable long-context inference, and robust tool calling.