Remote AI/ML Engineer for LLM Security & Detection

AuditSmart; United States; Remote; USD 120,000 - 180,000; Full time

Benefits offered by this job

Remote work

Job summary

AuditSmart seeks a remote, full-time AI engineer to design and improve a multi-agent pipeline for detecting software vulnerabilities, emphasizing prompt and agent-system design across hosted and custom language models.

You will evaluate AI outputs and collaborate with model providers such as Groq, Claude, Gemini, and fine-tuned models to guide improvements in accuracy and reliability. Proficiency in LangChain and LangGraph, plus evaluation frameworks, is essential.

Qualifications

• Strong background in LLM prompt engineering and agentic systems.
• Experience with LangChain, LangGraph, or a comparable framework.
• Familiarity with evaluation frameworks for AI accuracy.

Responsibilities

• Design and refine multi-agent workflows for vulnerability detection.
• Work with Groq, Claude, Gemini, and custom fine-tuned models.
• Apply evaluation methods to assess AI accuracy and guide improvements.

Skills

LLM prompt engineering

agentic systems

multi-agent workflows

AI evaluation

model evaluation

Tools

LangChain

LangGraph

Groq

Claude

Gemini

Job description

AuditSmart seeks a remote, full-time AI engineer to design and improve a multi-agent pipeline for detecting software vulnerabilities, emphasizing prompt and agent-system design across hosted and custom language models.

You will evaluate AI outputs and collaborate with model providers such as Groq, Claude, Gemini, and fine-tuned models to guide improvements in accuracy and reliability. Proficiency in LangChain and LangGraph, plus evaluation frameworks, is essential.