Agent Systems Engineer
Jobtailor
Responsibilities
- Design and implement long-running, harness-based multi-agent architectures — covering topologies, orchestration, memory management, tool registries, and execution environments.
- Develop behavioral and persona models with goal-directed simulations, including evaluation frameworks, prompt schemas, and automated or human-in-the-loop feedback mechanisms to measure fidelity and surface failure modes.
- Build signal and insight systems that translate agent outputs into actionable product decisions, enabling data-driven iteration across research, product, and engineering teams.
- Ensure production reliability through comprehensive tracing, cost monitoring, failure detection, and robust fallback strategies.
- Collaborate cross-functionally with distributed systems, machine intelligence, and product teams to deliver measurable, scalable agent capabilities.
Requirements
- 5+ years of hands-on experience designing and implementing production-grade distributed systems and agent-based architectures
- Strong proficiency in Python, with substantial async programming experience.
- Demonstrated experience building production ML systems, including tracing, cost monitoring, failure detection, and robust fallbacks.
- Hands-on experience with containerisation and orchestration (Docker, Kubernetes) for large-scale agent infrastructures.
- Experience designing behavioral models, goal-directed simulations, evaluation frameworks, prompt schemas, and human-in-the-loop feedback mechanisms.
- Proven ability to translate agent outputs into actionable product decisions; experience building signal and insight systems.
- Strong cross-functional communication and collaboration skills across technology, research, and product.
Core Competencies
Expertise in designing and implementing production-grade distributed systems and agent-based architectures, with a strong focus on Python programming, containerization, and orchestration. Proven ability to develop behavioral models and translate agent outputs into actionable insights for product decisions.
#J-18808-LjbffrReference: WJ-766_21912632