AI Engineer II

Lennar•Miami, FL

19d

About The Position

AI Engineer II We are Lennar Lennar is one of the nation’s leading homebuilders, dedicated to making an impact and creating an extraordinary experience for their Homeowners, Communities, and Associates by building quality homes and providing exceptional customer service, giving back to the communities in which we work and live in, and fostering a culture of opportunity and growth for our Associates throughout their career. Lennar has been recognized as a Fortune 500® company and consistently ranked among the top homebuilders in the United States. Join a Company that Empowers you to Build your Future We are seeking a technically grounded AI Agent Developer to design, deploy, and scale our internal agentic framework. Working within the LTG Corporate Analytics team, you will build the infrastructure that powers intelligent finance agents — translating enterprise data into automated, context-aware workflows. This role sits at the intersection of data engineering and AI orchestration: you will configure MCP servers, develop domain-specific subagents (beginning with Finance), and integrate them with our Snowflake data platform, dbt semantic layer and models, and AWS S3 infrastructure. This is not an LLM research role — it is a hands-on engineering position for someone who understands how enterprise data ecosystems work and wants to extend that expertise into the agentic layer. A career with purpose. A career built on making dreams come true. A career built on building zero defect homes, cost management, and adherence to schedules.

Requirements

Bachelor’s degree in Computer Science, Information Systems, Data Engineering, or a related technical field.
3–5+ years of experience in data engineering, analytics engineering, or backend development — with a strong understanding of how enterprise data ecosystems are structured and maintained.
Hands-on experience with Snowflake: query development, data modeling concepts, schema design, and performance tuning.
Proficiency with dbt: building and maintaining models, understanding the semantic layer and metrics definitions, and working within a structured transformation workflow.
Familiarity with AWS infrastructure, particularly S3: bucket organization, IAM access patterns, lifecycle policies, and integration with downstream consumers.
Strong SQL skills for querying and transforming complex enterprise datasets across multiple sources.
Proficiency in Python for scripting, API integration, and building lightweight data pipelines or agent tooling.
Conceptual understanding of agentic frameworks: tool use, orchestration, context management, and MCP or similar protocol patterns — no LLM training or ML research experience required.
Experience designing for production environments: reliability, monitoring, error handling, and performance optimization.
Ability to translate complex business requirements into clean, maintainable technical implementations.
Strong communication skills to partner effectively with finance stakeholders and explain technical designs to non-engineering audiences.

Nice To Haves

Experience with ERP systems (particularly JD Edwards / Oracle JDE) or other enterprise financial data sources.
Familiarity with Model Context Protocol (MCP) or hands-on experience building tool-callable APIs for AI agent consumption.
Prior exposure to agentic frameworks such as LangGraph, CrewAI, or similar orchestration tools — from an engineering and deployment perspective.
Experience working within a modern analytics stack (Snowflake + dbt + Airflow or equivalent) in a corporate data & analytics team.
Background in data governance: access controls, audit logging, lineage tracking, and schema documentation.
Familiarity with financial reporting concepts (GL, AP/AR, cost accounting) that are relevant to Finance Agent use cases.

Responsibilities

Design, configure, and scale MCP (Model Context Protocol) servers that expose Snowflake data, dbt metrics, and S3-stored assets as tool-callable resources for AI agents.
Build and maintain domain-specific subagents (e.g., Finance Agent) that consume the MCP layer to answer business questions, surface data insights, and automate reporting workflows.
Integrate agents with the dbt semantic layer and underlying models — ensuring semantic consistency between how metrics and dimensions are defined in dbt and how agents retrieve and present them.
Manage S3 bucket structure and access patterns that support agent context: storing prompt artifacts, retrieval corpora, output logs, and intermediate agent state.
Partner with data engineers and analytics leads to understand domain data models (JDE, Snowflake silver/gold layers) so agents are grounded in accurate, well-governed data.
Establish agent orchestration patterns: define tool invocation sequences, fallback logic, human-in-the-loop escalation rules, and multi-agent handoff protocols.
Build observability into the agent layer: logging, tracing, usage dashboards, and performance monitoring so agent reliability can be measured and improved over time.
Maintain governance and data access controls across all agent integrations — ensuring agents operate within approved data boundaries and produce auditable outputs.
Document agent configurations, data contracts, MCP server specs, workflow diagrams, and runbooks for other engineers onboarding to the platform.
Stay current with the rapidly evolving agentic tooling landscape and bring relevant patterns back to the team — with a focus on operational frameworks rather than model research.