Principal Full Stack Engineer

Lennar•Waterford, FL

About The Position

Lead the architecture and delivery of production-grade AI-powered systems across the full stack. Operate at the intersection of applied AI research and software craftsmanship - designing agentic workflows, integrating large language models, and establishing engineering practices at scale. Serve as the technical standard-bearer for how the organization builds, evaluates, and operates AI features. Partner with Engineering and Data to translate ambiguous business problems into concrete technical roadmaps. Drive AI engineering strategy across the stack, authoring RFCs and leading design reviews that raise the bar for the entire engineering org.

Requirements

8+ years of professional software engineering with at least 2 years building production AI / ML systems.
Expert-level proficiency in Go and TypeScript / JavaScript; comfortable owning the full stack from model to UI.
Hands-on experience designing and shipping LLM-powered features including prompt engineering, fine-tuning, and RLHF concepts.
Deep understanding of transformer architectures, token economics, context window management, and inference optimization.
Proven track record with RAG architectures, embedding models, and semantic search at scale.
Strong grasp of distributed systems, API design, and cloud-native infrastructure (AWS).
Experience with MCP (Model Context Protocol), tool-use patterns, and multi-agent orchestration in production.
Demonstrated ability to lead without formal authority, influencing direction through technical credibility and clear communication.

Nice To Haves

Contributions to open-source AI projects or published technical writing on AI engineering topics preferred.
Experience in regulated or enterprise environments where auditability, security, and compliance are first-class concerns is a plus.

Responsibilities

Design and own end-to-end AI features: from prompt engineering and model selection through to scalable APIs and responsive front-end surfaces.
Architect multi-agent and agentic workflow systems using frameworks such as LangGraph, CrewAI, AutoGen, or custom orchestration layers.
Build retrieval-augmented generation (RAG) pipelines with production-grade chunking, embedding, indexing, and reranking strategies.
Integrate LLM providers (OpenAI, Anthropic, Gemini, local models via Ollama / vLLM) with robust fallback, rate-limiting, and cost controls.
Deliver polished, accessible front-end experiences in React that expose AI capabilities with low latency and graceful degradation.
Build and maintain high-throughput back-end services in Node.js, or Go serving as the inference and orchestration layer.
Define and implement LLM evaluation frameworks evals, red-teaming, regression suites to continuously measure model quality in production.
Instrument AI pipelines with tracing, latency histograms, and cost dashboards using tools such as LangSmith, Helicone, or Grafana.
Embed responsible-AI practices: hallucination mitigation, PII redaction, output guardrails, and bias monitoring into every feature.
Mentor senior engineers, conduct rigorous code and architecture reviews, and contribute to hiring by defining role expectations and screening candidates.