Sr. Software Engineer – AI and Automation (Team Breeze)

Who Are We?

Taking care of our customers, our communities and each other. That’s the Travelers Promise. By honoring this commitment, we have maintained our reputation as one of the best property casualty insurers in the industry for over 170 years. Join us to discover a culture that is rooted in innovation and thrives on collaboration. Imagine loving what you do and where you do it.

Technology

Compensation Overview

The annual base salary range provided for this position is a nationwide market range and represents a broad range of salaries for this role across the country. The actual salary for this position will be determined by a number of factors, including the scope, complexity and location of the role; the skills, education, training, credentials and experience of the candidate; and other conditions of employment. As part of our comprehensive compensation and benefits program, employees are also eligible for performance-based cash incentive awards.

Salary Range

$142,900.00 - $235,800.00 1 At Travelers, we're building the future of insurance — and Team Breeze is at the center of that work. This role puts you at the intersection of large language models (LLMs), cloud-native infrastructure, and real business impact. If you've shipped AI agent systems to production and want to do it again at scale, we'd like to talk.

What Will You Do?

• Lead the design, configuration and development of complex, high-visibility AI applications on Team Breeze with a focus on LLM agents, retrieval pipelines and production reliability.
• Design and implement agent orchestration flows using AWS Bedrock AgentCore and/or frameworks such as LangGraph, including tool calling, routing and prompt strategies for Claude models on Bedrock.
• Own the retrieval-augmented generation (RAG) pipeline end to end — from document ingestion (SharePoint/LivePublish to S3), processing and embedding in Databricks, and serving from a vector store via Python retrieval APIs (FastAPI/Haystack) — including evaluation steps that measure retrieval and answer quality before each release.
• Architect and build services on AWS using Lambda, API Gateway, S3, CloudFront/Lambda@Edge, Route 53, Secrets Manager/Vault and Bedrock application inference profiles to track and manage cost.
• Develop and operate containerized retrieval and evaluation services on Amazon Elastic Kubernetes Service (EKS), deployed with Helm and ArgoCD, including OpenTelemetry collector integration exporting traces to Dynatrace, S3 and Braintrust.
• Implement and maintain infrastructure as code and CI/CD using Terraform (Terraform Enterprise), GitHub Actions and TEP pipelines to deliver reliable, repeatable deployments.
• Build and maintain the application layer across Python and TypeScript/Node 24 (Lambda xAPIs) and React 19/Vite for front-end UI experiences.
• Lead investigation and resolution of critical, high-impact production issues, including debugging across AWS services, Kubernetes workloads, data pipelines and observability tooling.
• Provide technical guidance and mentorship to teammates, fostering strong engineering practices around design documentation, testing, evaluations, code reviews and operational readiness.
• Continuously apply knowledge of current AI/LLM, cloud and observability trends to improve system design, developer productivity and business outcomes.
• Perform other duties as assigned.

What Will Our Ideal Candidate Have?

• LLM and Agent Systems
  • Demonstrated experience shipping LLM/agent systems to production — not just demos or prototypes.
  • Hands-on knowledge of tool calling, prompt and routing design, RAG retrieval quality tuning, and defining evaluations to measure LLM and retrieval performance.
  • Experience with Strands, LangGraph or Bedrock AgentCore is a plus.
• Cloud, Infrastructure and Operations
  • Strong AWS serverless and EKS experience; able to design and debug systems spanning Lambda, API Gateway and Kubernetes-based containers in production.
  • Practical experience authoring and maintaining Terraform, Helm charts and CI/CD pipelines (GitHub Actions, TEP).
  • Observability fluency: tracing with OpenTelemetry or Dynatrace, structured logging with correlation IDs, cost tracking and incident handling.
• Full-Stack and Data Pipeline Ownership
  • Comfortable working across Python and TypeScript/Node 24 in the service layer and React 19/Vite on the front end.
  • Familiar with data and retrieval pipelines from ingestion through embedding to vector-store-backed APIs.
  • Consistent habit of writing tests, evaluations and data-quality checks.
• Delivery
  • Advanced delivery skills, including the ability to determine software design strategy and methodology, use automated tests and informed feedback loops to ensure quality and production readiness, and monitor the health of work efforts and adjacent systems.
• Domain Expertise
  • Demonstrated track record of developing business partnerships and influencing priorities by identifying solutions aligned with current business objectives; closely follows industry trends relevant to the domain and shares knowledge with coworkers.
• Problem Solving
  • Strong problem solver who uses data and proofs of concept to find creative solutions, reflects on outcomes, measures impact and uses that information to optimize — including decisions that involve a significant number of factors with broad implications.
• Communication
  • Able to describe technology concepts in ways the business can understand, document initiatives clearly and concisely, collaborate effectively regardless of role, and give and receive constructive feedback; an attentive and empathetic listener.
• Leadership
  • Takes action even when there is no clear owner; inspires and motivates others and is effective at influencing team members.

What is a Must Have?

• Bachelor’s degree in computer science, related STEM field, or its equivalent in education and/or work experience.
• 6 additional years of software engineering experience.