H
Applied AI Engineer
HSF
3 hours ago
Full-time
On-site
Los Angeles, California, United States
Applied AI Engineer
The AI layer is the beating heart of the company. As our AI/LLM engineer, you won't just write prompts — you'll define how the system thinks. You'll architect the orchestration layer that enables streaming chat, multi-agent coordination, and backend execution across ERP workflows. Your work will determine how our software reasons, adapts, and collaborates with users — at scale, and in real time. What you'll do: Build agentic workflows that track state, propose actions, and complete multi-step tasks Integrate models with tools, memory, and retrieval over structured and unstructured data Protect sensitive information and build recovery behavior for failed tools, incomplete information, and uncertain model outputs Build evaluations that measure AI system quality and guide improvements Refine AI interactions for responsiveness, clarity, and user control Collaborate with frontend and backend engineers to bring AI into real user workflows Make thoughtful tradeoffs between model capability, latency, and reliability
The AI layer is the beating heart of the company. As our AI/LLM engineer, you won't just write prompts — you'll define how the system thinks. You'll architect the orchestration layer that enables streaming chat, multi-agent coordination, and backend execution across ERP workflows. Your work will determine how our software reasons, adapts, and collaborates with users — at scale, and in real time. What you'll do: Build agentic workflows that track state, propose actions, and complete multi-step tasks Integrate models with tools, memory, and retrieval over structured and unstructured data Protect sensitive information and build recovery behavior for failed tools, incomplete information, and uncertain model outputs Build evaluations that measure AI system quality and guide improvements Refine AI interactions for responsiveness, clarity, and user control Collaborate with frontend and backend engineers to bring AI into real user workflows Make thoughtful tradeoffs between model capability, latency, and reliability