AI Engineer

Role overview Build and operate the intelligence behind an AI assistant for compliance professionals. You will take language-model features from early design through production, developing multi-turn agents and the systems that measure their quality, speed, and cost. The work also includes enabling external AI agents to interact with the assistant’s capabilities securely and reliably.

Responsibilities - Own AI assistant capabilities from concept through implementation, production rollout, and ongoing operation. - Design multi-turn agents that can reason across steps, retain context, and take actions. - Build tracing, monitoring, and evaluation pipelines to track model quality, latency, and token costs. - Improve interfaces and capabilities that let external AI agents interact with the system securely and consistently. - Start with reliable baseline approaches, then refine complexity using production metrics and A/B tests.

Requirements - At least five years in software, data, or machine learning engineering, including at least one year building and shipping production GenAI or LLM systems. - Strong understanding of LLM behavior and practical techniques such as structured extraction, tool calling, and cost-aware model routing. - Hands-on experience with agent orchestration frameworks such as LangChain or LangGraph and modern LLM APIs. - Ability to build production microservices, with experience in Java or Dropwizard and cloud deployment using GCP or AWS, Docker, and Kubernetes. - Ownership mindset spanning customer discovery and production operations.

Nice to have Experience in NLP, MLOps, large-scale data engineering, or production retrieval-augmented generation; familiarity with compliance, fraud prevention, KYC/AML, or fintech.

Benefits and work setup Remote-first work with flexible hours and optional office or coworking access in some locations. Time off includes a birthday holiday, 10 personal days annually, seven sick days without paperwork, and additional time around Christmas and New Year. Compensation is market-benchmarked and varies by confirmed work location.