Principal / Staff Engineer AI Systems | Agentic AI & Knowledge Platforms

Position Overview / Ideal CandidateÉlan Partners is seeking Principal/Staff-level AI Engineers for two highly technical opportunities supporting the development of a large-scale, production AI platform. These roles are focused on building the core systems that enable autonomous AI agents to reason, make decisions, and safely take action within complex enterprise environments. We are seeking engineers specializing in one of two areas: Agentic AI Systems or AI Knowledge & Data Systems. Experience across both areas is not required. These are hands-on engineering roles for candidates who have moved beyond AI prototypes, proofs of concept, and chatbot applications and have experience designing, building, and owning AI systems in production at scale. The strongest candidates will have direct architecture ownership, experience solving difficult production problems, and a track record of building systems that are reliable, auditable, scalable, and trusted in real-world environments. Track 1 – Agentic AI SystemsThis role focuses on the architecture and production deployment of agentic and autonomous AI systems, including multi-agent orchestration, reasoning, planning, evaluation, guardrails, and defining the boundaries between autonomous execution and human oversight. Ideal experience includes:7–8 years of overall software engineering experience3–4 years of recent, hands-on experience building and shipping agentic or autonomous AI systems in production Experience architecting multi-agent systems and autonomous workflows Direct ownership of core AI architecture rather than solely contributing to an existing design Experience with LLMs and multi-model orchestration, including selecting/routing models based on reasoning, latency, or execution requirements Experience designing AI evaluation frameworks/evals to measure and validate agent behavior and reliability Experience defining guardrails, fallback mechanisms, human-in-the-loop controls, and autonomy/trust boundaries Strong understanding of production reliability, latency, observability, security, and governance for AI systems Ability to design systems where AI actions and the reasoning behind them are traceable and auditable The source role specifically emphasizes engineers who can determine when an agent should execute independently versus defer to human oversight and who can build deterministic safeguards around autonomous behavior. Track 2 – AI Knowledge & Data SystemsThis role focuses on building the knowledge and semantic data foundation that production AI agents depend on, including knowledge graphs, ontologies, semantic layers, entity resolution, and real-time data reconciliation. Ideal experience includes:7–8 years of overall engineering experience3–4 years of recent, hands-on experience building production knowledge graphs, ontologies, or semantic layers supporting live AI products Experience architecting and scaling knowledge graphs and semantic data platforms Strong experience with entity resolution and data orchestration Experience integrating, reconciling, and linking complex or conflicting data across multiple systemsExperience building low-latency data pipelines supporting real-time AI reasoning and decision-making Ability to manage dynamic state and maintain accurate semantic relationships as underlying data changes Experience taking data/AI platforms from prototype or demo to trusted production scale Strong understanding of data quality, reconciliation, reliability, observability, and governance The emphasis is not simply on data modeling. The role requires experience dealing with messy, conflicting real-world data and maintaining a reliable, real-time semantic representation that AI systems can use to make operational decisions. Experience That Will Stand OutFor either track, we are particularly interested in candidates who have: Built sophisticated AI systems within an AI-native, high-growth technology, or large-scale technology environment Taken direct end-to-end architectural ownership of production systemsExperience moving complex AI technology from "works in a demo” to trusted production use Worked within highly regulated, mission-critical, or enterprise environments Designed for security, governance, traceability, and auditability from the beginning rather than treating them as afterthoughts Significant experience diagnosing real production challenges involving latency, reliability, data quality/reconciliation, evaluation, and system behavior Both profiles place significant emphasis on engineers with real production experience who can discuss failures, tradeoffs, reliability problems, and what it took to make complex AI systems trustworthy at scale.