Senior Application Security Engineer, AI and Machine Learning
$180k - $220k • San Francisco, California, United States; Seattle, Washington, United States
Posted 1mo ago
Job Location
San Francisco, California, United States; Seattle, Washington, United States
Tech Stack
Remote Work Policy
On-site
Categories
AI Engineer
About the job
Lightning AI is seeking a Senior Application Security Engineer to safeguard their AI, machine learning, and inference platforms. This is a hands-on technical role focused on integrating security into modern AI infrastructure, inference systems, and developer platforms. You will collaborate with platform engineers, ML engineers, and infrastructure teams to identify risks, design secure architectures, and develop security tooling that empowers engineers to operate efficiently and securely. The role is execution-driven, involving technical implementation, in-depth security reviews, and contributing to the growth of application security capabilities.
Responsibilities
- Perform threat modeling across AI platforms, inference services, and ML pipelines.
- Identify risks such as prompt injection, model extraction, adversarial inputs, and data leakage.
- Review model serving architectures and inference pipelines.
- Partner with ML engineers to secure training, fine tuning, and deployment workflows.
- Design isolation and security controls for multi-tenant AI workloads.
- Perform architecture and design security reviews.
- Conduct targeted code reviews for high-risk components.
- Identify security gaps in APIs, micro-services, and distributed systems.
- Build secure patterns for authentication, authorization, and service-to-service communication.
- Help engineering teams implement secure defaults and guardrails.
- Secure customer-facing inference APIs and services.
- Protect against abuse, model extraction, and adversarial behavior.
- Design rate limiting, isolation, and workload protection controls.
- Build monitoring and detection for anomalous inference behavior.
- Evaluate open-source models and dependencies.
- Secure model artifacts and distribution pipelines.
- Implement integrity validation and provenance controls.
- Secure container images and runtime environments.
- Build security automation and tooling.