Principal Agentic Software Engineer

arrivia. Go far in the travel industry.•Scottsdale, AZ

About The Position

As a Principal Agentic Software Engineer, you are a hands-on, product-minded, customer-focused full-stack engineer who sets technical direction across multiple products and platforms and personally builds the primitives other teams depend on. You still open the editor yourself, and you are a driving force behind how agentic development works at arrivia. You do not advise from the sidelines. You direct AI coding agents through real, multi-step work, decide what becomes shared infrastructure, and make the build-vs-buy calls that shape our roadmap. Operating with broad autonomy and cross-organizational accountability, you own multiple domains or a critical platform. Your work reaches millions of people. We build the travel and rewards platforms behind major banks, membership organizations, and travel brands, where members book real trips with real money and points, so trust is the product. The operating rhythms, launch gates, and quality standards you set let dependent teams move faster while protecting that trust. You will balance short-term partner requests against long-term member trust, lead the one-way-door decisions in your areas, and define how AI coding agents reshape the way we build and operate at scale. You work across many technology domains, business teams, and offices in a fast-paced, highly collaborative environment.

Requirements

A recent, hands-on track record. You have personally built and shipped working software in the last year or two, not only led teams who did.
8 to 12 years of full-stack development and architecture experience building web applications and services, including technical leadership beyond a single team.
A bachelor's degree in Computer Science, Computer Engineering, or equivalent experience.
Advanced, hands-on use of AI coding agents such as Claude Code, Cursor, or GitHub Copilot, including directing them through larger multi-step work, with a track record of building practices that others adopt.
Strong proficiency in two or more of TypeScript/JavaScript, Python, and C#/.NET, plus modern frontend frameworks such as React, Next.js, Vue, or Angular.
Deep experience with RESTful API design, distributed systems, and microservices at scale, and cloud-native delivery on AWS, Azure, or GCP including multi-region or hybrid deployments.
Hands-on depth with the Model Context Protocol, prompt engineering for code, and AI building blocks such as Azure AI Foundry, RAG, LangChain, or graph-based data technologies like GraphRAG and GraphDB.
Production experience with real-time streaming and distributed messaging such as Redis, Kafka, or RabbitMQ, plus containerization and orchestration with Docker and Kubernetes, advanced CI/CD, infrastructure-as-code, and modern DevOps or SRE practices.
A record of mentoring senior engineers and raising the bar through frameworks and standards, and of explaining complex AI and architectural ideas simply to executives and non-technical partners.

Responsibilities

Set technical direction across multiple products and platforms, including AI infrastructure, data platforms, and shared primitives, and personally build the components that increase leverage for teams beyond your own.
Co-define our prompt libraries, MCP server patterns, quality gates for AI-generated code, and human-agent collaboration patterns, then get them adopted.
Architect Model Context Protocol integrations that give AI agents secure, governed access to internal APIs, data sources, and platform services across teams.
Drive build-vs-buy decisions and convergence on strategic standards that reduce complexity rather than add to it.
Institutionalize prioritization mechanisms, AI launch checklists, and SLOs that raise quality and speed for dependent teams.
Lead one-way-door decisions in your areas, minimize the chance of failure, and hold contingency plans for failure scenarios.
Guide teams on what is worth pursuing and what to kill, and define what success looks like for your domains.
Standardize canonical metrics, AI eval frameworks, minimum sample sizes, and guardrail dashboards, and hold teams to them before anything scales.
Co-define responsible-AI guardrails and review processes so high-risk use cases get scrutiny and post-incident learnings feed back into design standards.
Turn legacy systems into scalable, cloud-native, agent-ready services through clear migration paths and architectural patterns.
Safeguard members' sensitive data by defining security patterns and partnering with our Risk, Security, and Compliance teams.
Mentor staff-level engineers and shape technical-leadership norms across individual contributors, so strong solutions emerge from teams and not from you alone.