Principal Full Stack Engineer

Lennar•Miami, FL

About The Position

We are looking for a Principal Engineer who will fundamentally change how our engineering teams work. Transforming the internal culture, workflows, and platform of our engineering organization to be AI-native. You will be the person who embeds with teams, rolls up your sleeves, and shows engineers firsthand how tools like Claude Code, ChatGPT Codex, and GitHub Copilot paired with agentic CI/CD patterns can compress days of work into hours. You will simultaneously build the platform scaffolding that makes this transformation repeatable, safe, and scalable across the organization. This is a builder and coach role in equal measure and the ideal candidate thrives in both. 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

8+ years of engineering experience with strong foundations in backend services, APIs, cloud-native infrastructure, and containerized deployments.
Deep CI/CD expertise - GitHub Actions, ArgoCD with hands-on experience redesigning pipelines to incorporate AI-driven steps, agent checkpoints, and human-in-the-loop approval gates.
Experience in platform engineering, internal developer platforms (IDPs) - you understand that engineers are your customers.
Fluent in GoLang and TypeScript/JavaScript; comfortable writing the glue code, automation scripts, and workflow tooling that makes platforms hum.
Direct, hands-on experience with AI coding tools in production engineering contexts - Claude Code, GitHub Copilot, or equivalent - not just personal use, but driving adoption with others.
Practical knowledge of LLM integration patterns: structured prompt engineering, tool/function calling, MCP server design, agentic frameworks (LangChain, LlamaIndex, or custom), context management strategies, and multi-step reasoning workflows with defined failure modes.
Deep understanding of agentic system design - how to architect multi-step agent workflows with tool use, memory, sandboxed execution environments, human-in-the-loop checkpoints, rollback mechanisms, and token budget controls that prevent runaway costs or unintended side effects.
Experience evaluating AI tooling for organizational fit: model capability benchmarking, security and data residency posture, SSO/SCIM integration requirements, rate limit and cost modeling at scale, and shadow IT risk when teams self-adopt without guardrails.

Responsibilities

Embed directly with engineering teams to drive hands-on adoption of AI coding tools - Claude Code, ChatGPT Codex, GitHub Copilot going beyond demos to changing actual daily habits: AI-assisted code review, context-aware test generation, and agent-driven refactoring workflows.
Design and run structured transformation programs: onboarding sprints, office hours, lunch-and-learns, and pairing sessions that meet engineers where they are.
Build and maintain the internal AI transformation playbook - documenting CLAUDE.md conventions, MCP server configurations, agent tool definitions, prompt libraries, and workflow templates that teams can self-serve and adapt.
Partner with engineering managers and tech leads to identify the highest-leverage transformation opportunities within each team's existing workflows.
Redesign CI/CD pipelines to incorporate agentic steps: automated code review agents, AI-driven test generation, PR summarization, release note automation, and self-healing build workflows.
Build the internal platform layer that makes AI tooling safe and repeatable at scale - guardrails, cost controls, audit logging, prompt governance, and approval workflows for agentic actions.
Define and implement agentic patterns for platform engineering: IaC agents that auto-remediate drift, incident response agents triggered by Coralogix/Grafana alerts, Claude Code sub-agents for deployment validation, and multi-step orchestration across GitHub Actions and internal tooling.
Architect integrations between AI tooling and the existing engineering ecosystem: GitHub, Jira, Slack, internal developer portal, observability platforms, and secrets/identity systems - ensuring agents operate within defined permission boundaries.
Own the technical strategy for AI tooling adoption - model selection across frontier providers (Anthropic, OpenAI, open-source), build vs. buy decisions for agentic infrastructure, token cost management, context window optimization, and responsible deployment practices across the engineering org.
Establish evaluation frameworks for agentic workflows: LLM-as-judge harnesses, task completion rate benchmarks, hallucination and tool-misuse detection, latency/cost tradeoff analysis, and DORA metric correlation to AI adoption milestones.
Mentor and elevate senior engineers by modeling AI-native engineering practices - through design reviews, documentation, pairing, and shared tooling.
Contribute to cross-functional AI governance conversations, ensuring our agentic platform practices align with security, compliance, and organizational risk tolerance.