Forward Deployed Engineer & Agentic Workflow Engineer – AI Innovation & Transformation
Summary
Build and deploy production-grade agentic AI workflows for finance teams, integrating LLMs with ERP/CRM systems and mentoring clients on rapid AI adoption.
What you'll do:
- Rapid Jumpstart deployment: Deploy AI Jumpstart packages within client environments, configuring data connectors, calibrating agent logic, and validating outputs against source-system controls within the first week of engagement.
- Agentic workflow design & build: Architect and implement multi-agent orchestration workflows using frameworks such as LangGraph, CrewAI, or AutoGen; design task decomposition, inter-agent messaging, tool-call schemas, and human-in-the-loop (HITL) / human-on-the-loop (HOTL) checkpoints appropriate to the risk and materiality of each workflow step.
- LLM integration & prompt engineering: Write, version, and optimize system prompts and structured output schemas for LLM-powered agents; implement function-calling / tool-use patterns against financial APIs; and fine-tune retrieval strategies (hybrid structured + RAG) to maximize accuracy and auditability.
- Enterprise system integration: Build and certify MCP connectors and REST/GraphQL integrations to ERP (NetSuite, SAP, Oracle), EPM (Adaptive Planning, Anaplan), CRM (Salesforce), and HRIS (Workday) systems, ensuring data freshness, completeness, and reconciliation to source-system controls.
- Production deployment & reliability: Package agents as containerized microservices (Docker/Kubernetes or equivalent) and configure CI/CD pipelines, environment promotion (dev → staging → production), and monitoring/alerting so agents run reliably at client scale.
- Client co-development: Work shoulder-to-shoulder with client finance, IT, and data teams; translate business requirements into technical agent design; facilitate working sessions; demonstrate working agents to executive stakeholders; and iterate rapidly based on feedback.
- SOX & audit-readiness: Instrument agents with deterministic logging, source citations, and control documentation so every agent-generated output can be traced, validated, and presented to internal or external auditors.
- Contribute to developing and implementing firm-approved, AI-enabled solutions for clients, in accordance with company policies on data protection, intellectual property, and professional standards.
- Stay informed about emerging AI tools and techniques and collaborate with firm leadership to identify compliant opportunities to enhance client solutions and internal processes.
Practice Leadership: Serve as a key leader in the AI Innovation & Transformation practice by:
- Developing reusable accelerators—contributing battle-tested code, workflow templates, and connector certifications back to the AI practice accelerator library to reduce deployment time on future engagements.
- Creating new delivery methodologies and service offerings that scale agentic solutions across clients and enterprise functions.
- Mentoring analysts and junior engineers on engagement teams, tracking and directing performance against objectives while encouraging continuous improvement and innovation.
- Contributing to recruiting, proposal writing, and firm-wide AI innovation initiatives.
What you'll bring:
- 5+ years of software engineering experience with at least 2 years focused on AI/LLM application development, agentic systems, or intelligent automation; prior “forward deployed” or client-embedded engineering experience strongly preferred.
- Hands-on production experience building multi-agent systems with LangChain/LangGraph, CrewAI, AutoGen, or equivalent frameworks; understanding of agent-loop design, task planning, tool-use, and memory management.
- Proficiency in Python (primary) and TypeScript/JavaScript; experience with FastAPI or equivalent for exposing agent capabilities as APIs.
- Deep understanding of LLM capabilities and limitations: prompt engineering, structured outputs, function calling, context-window management, cost optimization, and latency tradeoffs across frontier models.
- Experience integrating enterprise source systems via APIs and MCP connectors; familiarity with ERP, EPM, EDW/Data Lakes, and CRM data schemas in a finance context is a significant plus.
- Working knowledge of cloud data platforms (Azure, GCP, or AWS) and containerization (Docker, Kubernetes); experience with vector databases (Pinecone, Weaviate, or equivalent) and RAG pipelines.
- Finance domain fluency—ability to understand and implement workflows covering month-end close, variance analysis, revenue recognition, AR/AP, and FP&A without requiring extensive hand-holding from client finance teams.
- Exceptional ability to operate in ambiguous, fast-moving client environments; a demonstrated track record of delivering working software in compressed, high-stakes timelines.
- Clear, confident communication with both technical and non-technical stakeholders; ability to present live-agent demonstrations to CFOs and finance leadership.
- Continuous Learning Mindset: Openness to continuously learning and applying emerging LLM capabilities, agent frameworks, and enterprise integration patterns.
Qualifications:
- A bachelor’s degree from an accredited university in computer science, software engineering,
mathematics, or a related technical discipline. - Relevant certifications in cloud platforms (Azure AI Engineer, AWS Machine Learning Specialty,
GCP Professional ML Engineer), LangChain, or equivalent agentic AI tooling preferred. - Willingness to travel domestically up to 20%–40% (varies by client engagement phase).
- Availability to work on client site or in office 3 days a week, with 2 days remote (hybrid
environment).