Get Started
Hunting for you, 24x7Get the iPhone app

Your hunt runs in your pocket, and tells you the moment a strong match opens.

Download on theApp Store
OpenTrain AIPosted 12 days ago

GenAI Security Evaluation Engineer

Remote · WorldwideRemoteContract
Who we'd reach at OpenTrain AIFound and reached by Power Apply
Best leadEngineering lead

Leads the team you would join

CTO

Signs off on engineering hires

Technical recruiter

Owns the opening

Power Apply

From your own Gmail and LinkedIn.

About the role

This is a remote, part-time contractor role open worldwide, for an application security specialist who will build and assess security tasks for generative AI systems. The work involves producing vulnerable and patched implementations of agent and retrieval-augmented generation applications, defining authorization contexts, and writing tests that confirm whether a patch works. Candidates need at least 20 hours a week and clear technical English. The role sits within the Generative AI and RLHF work of OpenTrain, a platform for AI training work.

What you’ll do

  • Create and assess security tasks for agent and RAG-based generative AI applications.
  • Build vulnerable implementations and their patched counterparts, then show the difference between them.
  • Follow data and action flows across files, callbacks, registries, and background jobs.
  • Define authorization contexts covering actors, trust boundaries, tenants, permitted data and actions, purposes, recipients, and document access.
  • Record findings with exact code spans, traps, control status, and remediation advice.
  • Evaluate sensitive information disclosure, excessive agency, prompt injection, tool abuse, and logic or authorization flaws.
  • Use a live model to show relevant paths where practical.

What they’re looking for

  • At least 3 years in application or product security, secure code review, penetration testing, or vulnerability research.
  • Data-flow reasoning, plus knowledge of authentication, authorization, least privilege, and approval flows.
  • Production coding experience.
  • Knowledge of the OWASP Top 10 for LLM Applications.
  • Clear technical English.

Nice to have

  • CWE mapping, exploit and reproduction tests, CodeQL or Semgrep, and SARIF.
  • LLM tool calling, agents, MCP, RAG, memory, and tracing, with LangChain, LlamaIndex, relevant SDKs, and OpenTelemetry.
  • Docker and Git.
  • SWE, OSCP, GWAPT, or CSSLP credentials, CVE credits, bug bounty work, or published research.
  • AI red teaming or LLM evaluation experience.

What’s on offer

  • Part-time contract of 20 or more hours a week.
  • Remote, open worldwide.
  • The listing does not state a pay rate.

Skills

Questions about this role

Is this role remote?

Yes, it is fully remote and open worldwide.

How many hours a week?

At least 20 hours a week, on a part-time contract.

How much experience is needed?

At least 3 years in application or product security, secure code review, penetration testing, or vulnerability research.