Agent Felet & Multi Agent Systems (Horizon AI)
Core
Designing and building the reasoning layer for a fleet of AI agents that diagnose GPU infrastructure problems, decide on actions, and explain reasoning to enable autonomous data center operations.
Role type
Principal to Distinguished Multi-Agent Systems Architect
Builds
Autonomous agent fleets for data center infrastructure management
Domain
AI/ML, Infrastructure Automation, Data Centers
Deliverable
production ML models
Required skills
LLM-based agentic systems design, agent behavior evaluation, autonomy SLO definition, multi-agent orchestration, tool/function calling at scale, LLM tracing, online evaluation, Python systems engineering, task decomposition, escalation logic, governance boundaries
Preferred skills
Operational/SRE background, human-in-the-loop system design, retrieval over operational data, defining autonomous system permissions
