Applied AI Engineer, Agentic Systems

Role overview Build and operate autonomous AI systems for clients in regulated and other high-consequence fields, from early prototypes through production deployment. The work includes designing bounded agent behavior, safe remediation, monitoring, and escalation, with responsibility for how systems perform after release.

Responsibilities - Design, build, and operate agentic systems for clients in fields including finance, security, healthcare, telecommunications, and government. - Create remediation workflows that assess findings and potential impact, simulate changes, apply them within agreed limits, and record the outcome. - Include monitoring, logging, and human escalation paths in system designs. - Work with assurance specialists to make systems verifiable against their intended constraints. - Own deployments end to end and investigate unexpected behavior with clients.

Requirements - Production experience building and operating autonomous or agentic systems that act on real environments without continuous supervision. - Strong distributed systems fundamentals, including handling partial failures, timeouts, retries, and unexpected responses. - Ability to work directly with regulated clients and explain technical decisions in clear language. - Experience with LLM orchestration frameworks and tool-calling patterns, including understanding their failure modes. - A habit of defining and enforcing an agent’s boundaries before building the actions it can take.