AI Systems Engineer

Description

The company makes browser security software for human and AI agent workforces. Its platform stops threats from the web, documents and email before they reach a user, and a newer product extends that protection to the AI agents that work alongside employees.

Fortune 500 companies and government agencies use the platform, and the company is backed by top-tier investors.

This role leads the platform and backend design behind the company's real-time security detection services. The work covers low-latency serving for small language models, prompt injection engines and content or intent classifiers across GPU and cloud clusters. You will work closely with ML researchers, but your main focus is backend services, GPU replica scaling, guardrails and automated evaluation.

Responsibilities

• Design, build and tune production serving pipelines for small language models and classical ML models, using quantization, batching and GPU acceleration.
• Own and grow backend microservices that deliver real-time threat detection, such as sds-service-prompt-injection and sds-service-ddr.
• Build usage-based monitoring, per-tenant rate limiting and runtime guardrails.
• Run GPU cluster replicas on Kubernetes and pod infrastructure, cutting cold-start time and controlling infrastructure cost.
• Build continuous evaluation, automated retraining, model CI/CD and regression gates that check detection quality against live benchmarks.
• Refactor legacy service boundaries, service-to-service authentication and backend database setups.

Requirements

• 5+ years of software engineering experience building distributed backend systems or high-throughput microservices.
• A proven record optimizing model inference with tools such as TensorRT, vLLM, ONNX Runtime or Triton Inference Server.
• Strong programming skills in Python, Go or C++.
• Hands-on experience with Kubernetes, container orchestration, GPU scheduling and infrastructure auto-scaling.
• Solid knowledge of CI/CD pipelines, MLOps or LLMOps practices and cloud infrastructure on AWS or GCP.
• Nice to have: experience building usage-based rate limiting or multi-tenant guardrail architectures.
• Nice to have: familiarity with database refactoring, authentication protocols and paying down backend technical debt.
• Nice to have: a background in cybersecurity, threat detection engines or DLP and PII filtering platforms.
• Nice to have: experience comparing runtime and accuracy trade-offs between local small language models and cloud LLMs such as Vertex AI.
• Nice to have: experience with AWS, GCP, Terraform, Kubernetes and Linux.

Conditions

Recruitment agencies must work through the company's applicant tracking system and be assigned to a specific requisition by the Talent team under a valid services agreement; resumes sent any other way become company property and are not eligible for a fee.