What are the responsibilities and job description for the Sr. Principal Agentic Engineer position at Monarch AI?
Sr. Principal Agentic Engineer, Multi-Agent Platform Monarch AI — Remote
Very competitive salary sign-on bonus product incentive bonus full benefits life insurance
About Monarch AI
Monarch AI is a vertical AI company building transformative, industry-specific multi-agent AI products for enterprise and middle-market companies — deeply integrated into the workflows and data systems that power each industry. Currently in stealth mode.
The Role
We're looking for an exceptional, hands-on AI engineer who has personally built and shipped production-grade AI agents used by real customers — ideally in enterprise AI, vertical AI, or at a leading AI-native company.
You'll be a major part of building Monarch's multi-agent platform - LLM and a set of agentic product systems that work across the systems that run a large enterprise — ERP, CRM, CPQ/quoting, to name a few (more revealed during interview)
We're looking for someone with deep experience building reliable Multi-Agent LLM orchestration, RAG and agentic retrieval systems, harness, multi-agent orchestration, tool/function calling, MCP/secure data connections, agent memory, evals, observability, long-running workflows, enterprise integrations, and production-grade agent infrastructure.
What We're Looking For
- You've personally built and shipped production-grade AI agents, either Enterprise or Middle Market
- Deep experience designing multi-agent systems: planning, implementing the coe (very hands on coder), orchestration, handoffs, state management, long-running workflows, failure recovery, human-in-the-loop approvals
- Sophisticated RAG and agentic retrieval across structured and unstructured enterprise data: embeddings, hybrid search, reranking, grounding, context engineering, retrieval evaluation
- Agents that reason over relational databases and data warehouses — semantic layers, NL-to-query, grounding answers to authoritative records
- Tool/function calling, MCP, APIs, OAuth, and reliable write actions into enterprise systems — agents that read from one system, reason across several, and act safely in another
- Enterprise CRM and data integration experience — Salesforce preferred, not required; strong fundamentals in connectors, pipelines, and cross-system identity/entity resolution
- Agent memory and context systems for durable, long-running sessions; event-driven agents that respond to data changes rather than waiting for a prompt
- Production engineering rigor: queues, concurrency, async jobs, idempotency, retries, rate limits, fallbacks
- Rigorous agent evaluation and observability: task-success metrics, regression suites, tool-call accuracy, trajectory debugging, hallucination measurement
- Strong Python and/or TypeScript; you own systems from architecture through deployment
- Secure enterprise AI design: multi-tenant isolation, RBAC, credential custody, audit logs, approval boundaries
- First-principles systems reasoning and strong product instincts — you care whether the agent completes the business objective, not just whether the output sounds plausible
The Profile
We're especially interested in AI engineers from leading vertical-AI, enterprise-AI, or agent-native companies, and ho have strong educational credentials, namely, Master's in AI/ML, CS, applied math, statistics, or an adjacent field strongly preferred; PhD a plus.
Exceptional production-grade experience building AI agents and agentic product platforms outweighs credentials
We look forwar to hearing from you!