AI Agent Engineer — Scale Safe, Smart Agents

TRM Labs; United States; Hybrid; USD 200,000 - 275,000; Full time

Benefits offered by this job

Equity plan

Job summary

TRM Labs is hiring for an AI Engineer to accelerate next-generation AI applications with a focus on LLMs and agentic systems. You will build robust pipelines, high-performance infrastructure, and tooling to deploy AI safely and at scale, serving investigators and analysts.

You will design and optimize agentic workflows, integrate with OpenAI/Anthropic or local models, and contribute to a culture of rapid experimentation and ownership in a security-focused environment.

Qualifications

• Strong engineering background with backend/system experience, Python preferred.
• Hands-on work building with LLMs, agents, and tooling frameworks.
• Comfort with agentic pipelines and optimizing information flow into AI systems.
• Focus on safety, scalability, and explainability in system design.
• Customer/user empathy for analysts and a bias toward rapid experimentation.

Responsibilities

• Architect and implement a robust agentic framework supporting tool use, context retrieval, memory, and planning.
• Build intelligent modular agents automating investigative tasks and aiding analyst decisions.
• Scale LLM infrastructure, including prompt engineering, RAG, and evaluation loops.
• Design safe, observable, auditable agent behaviors for high-sensitivity environments.
• Evaluate metrics like reasoning, latency, and hallucination while iterating from feedback.

Skills

Backend engineering

Systems thinking

Python

Agentic AI

LLMs

Tools

LangChain

vector DBs

prompt engineering

RAG

OpenAI/Anthropic models

Job description

TRM Labs is hiring for an AI Engineer to accelerate next-generation AI applications with a focus on LLMs and agentic systems. You will build robust pipelines, high-performance infrastructure, and tooling to deploy AI safely and at scale, serving investigators and analysts.

You will design and optimize agentic workflows, integrate with OpenAI/Anthropic or local models, and contribute to a culture of rapid experimentation and ownership in a security-focused environment.