TRM Labs
AI Agent Engineer
United StatesFull-timeGlobal
š° USD 200,000 - 275,000/yr
š Remote
Job Description
[AI-summarized by JobStash]
You will architect and implement robust agentic frameworks that support tool use, context retrieval, memory, and planning. You will build modular agents to automate investigative tasks and augment analyst decision-making. You will extend and scale LLM infrastructure, including prompt engineering, retrieval augmented generation, model serving, and evaluation loops. You will design safe, observable, and auditable agent behaviors and evaluate performance across metrics like reasoning, latency, success rate, and hallucination, iterating based on user feedback and telemetry.
Requirements
- āStrong engineering background with deep experience in backend or systems work (Python preferred)
- āHands-on experience building with LLMs, agents, and tooling frameworks such as LangChain, semantic caches, and vector databases
- āComfort working with agentic pipelines and optimizing information flow into AI systems
- āThoughtful approach to system design with emphasis on safety, scalability, and explainability
- āHigh product empathy and a bias toward experimentation and iteration
- āExperience with knowledge graphs, task orchestration, or AI safety a plus
Responsibilities
- āArchitect and implement a robust agentic framework that supports tool use, context retrieval, memory, and planning
- āBuild intelligent, modular agents that automate investigative tasks and augment analyst decision-making
- āExtend and scale LLM infrastructure including prompt engineering, RAG, and evaluation loops
- āDesign safe, observable, and auditable agent behaviors ensuring reliability in high-sensitivity environments
- āEvaluate performance across metrics like reasoning, latency, success rate, and hallucination and iterate based on feedback and telemetry
- āContribute to rapid experimentation and ethical AI deployment
Benefits & Perks
- āEquity plan eligibility
Tech Stack
Vector databaseorchestrationPythonobservabilityTelemetrysystemsLLMagentSemantic cacheTask orchestration