Product Security Engineer

Adobe Adobe · Enterprise · New York, NY +2

Mid-level engineer to build AI-powered security analysis platforms using LLM integrations, RAG, and vector databases. Develops end-to-end platforms and evaluates AI outputs, addressing AI-specific risks.

What you'd actually do

  1. Build security analysis capabilities using LLM integrations with Azure OpenAI, prompt engineering, retrieval-augmented generation, and vector-based context retrieval.
  2. Develop and maintain platforms end to end using React, Python FastAPI, Celery, Postgres, Redis, and Kubernetes with Argo.
  3. Evaluate LLM and retrieval outputs to ensure accuracy and reliability for internal users.
  4. Address AI-specific risks like prompt injection, data exposure, and output manipulation.
  5. Make architecture decisions for new security capabilities, balancing performance, scalability, maintainability, and responsible AI use.

Skills

Required

  • Python
  • JavaScript
  • React
  • AI systems
  • LLMs
  • prompt engineering
  • Azure OpenAI
  • vector databases
  • retrieval-based systems
  • AI-specific risks
  • evaluating AI outputs
  • cloud platforms
  • Azure
  • containerized deployments
  • CI/CD pipelines
  • Git
  • modern development workflows
  • Secure SDLC practices
  • application security
  • threat modeling

Nice to have

  • React experience
  • full-stack or backend systems

What the JD emphasized

  • AI-powered platforms
  • LLM integrations
  • retrieval-augmented generation
  • vector-based context retrieval
  • AI-specific risks
  • prompt injection
  • data exposure
  • output manipulation
  • evaluating AI outputs

Other signals

  • AI-powered platforms
  • LLM integrations
  • retrieval-augmented generation
  • vector-based context retrieval
  • AI-specific risks