Agentic AI Developer

West Monroe is seeking an Agentic AI Developer to design, build, and deploy production-grade autonomous agent systems that drive measurable business outcomes. This role goes beyond prompt optimization—focusing on architecting scalable, secure, and observable agent frameworks that integrate with enterprise systems and operate reliably at scale.

The ideal candidate combines strong software engineering capability with deep understanding of LLM behavior, tool orchestration, and AI system design. You will translate business requirements into structured, testable agent solutions moving outputs from “it works sometimes” to “it works reliably at scale.”

Key Responsibilities

• Agent Development & Integration
  • Design and implement production-grade autonomous and semi-autonomous agent systems (planner–executor, supervisor–worker, tool-using agents).
  • Architect multi-step reasoning workflows with clear state management, memory strategies, retry logic, and fallback mechanisms.
  • Develop and integrate tool/function calling patterns that connect agents to enterprise APIs, databases, CRMs, and internal systems.
  • Ensure systems are production-ready with logging, monitoring, CI/CD integration, and automated testing.
• Evaluation & Governance
  • Establish success criteria and metrics for agent/prompt performance (accuracy/relevance, reliability, latency, cost, user trust) and use them to drive iteration.
  • Build evaluation harnesses and automated regression pipelines; maintain test sets with edge cases/adversarial/role variants and run controlled experiments (including A/B tests).
  • Monitor runtime behavior (token usage, latency, failure modes, drift) and outputs for bias/safety/compliance; implement guardrails for hallucinations, prompt injection, tool misuse, and data leakage (including structured-output validation).
• Cross-Functional Delivery
  • Work with business stakeholders to translate requirements into effective agent designs and capabilities (goals, constraints, success criteria, and operating boundaries).
  • Partner with IT product owners, platform administrators, and developers to integrate and deploy agents into AI-enabled tools, workflows, and applications (including tool connections and data access patterns).
  • Communicate system behavior, trade-offs, and recommendations clearly to technical and non-technical audiences, including risk, reliability, cost, and operational impacts.
  • Support releases by producing agent documentation and adoption materials (behavior/spec changes, tool schemas, runbooks, change notes) and advising teams on standardized agent patterns and evaluation practices.
• Research and Continuous Improvement
  • Stay current on advancements in generative AI, agent architectures, orchestration frameworks, and evaluation methodologies.
  • Assess emerging tools for agent lifecycle management, observability, and structured output control.
  • Contribute to internal standards for agent development, testing, and governance.
  • Provide guidance to teams adopting agent-based automation patterns.

Qualifications

• Technical and Analytical Expertise:
  • Strong experience building agentic solutions in enterprise settings, including configured agents (prompt-first, policy/knowledge/tools configured) and orchestrated agents (coded workflows with tools, state, and routing).
  • Familiarity with MCP-style tool exposure or other governed approaches to safely connect LLMs to enterprise tools.
  • Proficiency in Python (TypeScript a plus) and comfort working with APIs, au