Software Development Engineer III - Back End, Genai Platform

Expedia Expedia · Hospitality · IL

Expedia is seeking a Software Development Engineer III to join their GenAI Platform team. This role involves designing, building, and operating backend services and APIs that power the company's shared generative AI platform. Key responsibilities include implementing RAG pipelines, integrating with LLM providers and vector stores, contributing to platform guardrails and safety features, and ensuring the responsible integration of AI/ML solutions. The role requires strong Python and FastAPI skills, experience with cloud environments (AWS), and familiarity with data stores like PostgreSQL and Elasticsearch. Prior experience with LLM systems, RAG, vector stores, and AI/ML safety practices is preferred.

What you'd actually do

  1. Design, build, test, and operate Python- and FastAPI-based backend services and APIs that power GenAI Platform capabilities, including RAG Manager, RAG Data Plane, and vector store services.
  2. Implement sound system design (including low-level design), API design, and data modeling for backend services that are reliable, secure, and easy for internal customers to adopt.
  3. Integrate backend services with LLM providers, Elasticsearch-based vector stores, and Aurora PostgreSQL as part of Expedia Group’s RAG and GenAI platform stack, ensuring robust and observable data and inference flows.
  4. Contribute to and enhance platform guardrails and safety features such as authentication, API key management, workload-to-workload authorization, evaluation hooks, and monitoring to keep GenAI applications responsible and compliant.
  5. Safely integrate and operate AI/ML‑enabled solutions, tools, and workflows that improve developer and traveler outcomes, including familiarity with AI‑driven systems and applying foundational AI/ML concepts to real-world platform products.

Skills

Required

  • Bachelor’s degree in Computer Science or a related technical field; or Equivalent related professional experience.
  • 5+ years of relevant professional experience.
  • Proven experience building and operating backend APIs and services in a cloud environment, preferably AWS, including automated testing, CI/CD pipelines, monitoring, logging, and participation in on-call rotations.
  • Strong proficiency in Python and experience with FastAPI or similar web frameworks (such as Flask, Django, or other asynchronous frameworks), with solid understanding of software design principles, data structures, and design patterns.
  • Hands-on experience with at least one data store technology (relational, NoSQL, or search), with familiarity using systems like PostgreSQL or Elasticsearch to support service-level data modeling and persistence.

Nice to have

  • Experience designing and operating internal platforms, shared services, or SDKs that are consumed by multiple teams, including clear technical documentation (design docs, runbooks, integration guides) and data-driven iteration.
  • Prior exposure to or hands-on work with LLM-powered systems and GenAI concepts such as prompting, embeddings, retrieval-augmented generation (RAG), vector stores, and associated evaluation or observability frameworks.
  • Demonstrated ability to apply AI/ML and GenAI patterns safely within production services, including implementing and evolving guardrails, evaluations, and operational practices for AI/ML‑enabled solutions at scale.
  • Experience with RAG or vector store tooling (for example, LangChain, LlamaIndex, LangGraph) and familiarity with workflow or agent orchestration systems (such as n8n, Flowise, or similar) to integrate AI-driven capabilities into backend service architectures.
  • Background working with multiple programming languages and understanding their trade-offs for different backend and data-intensive workloads, with a track record of making pragmatic, data-informed technical decisions.

What the JD emphasized

  • production code
  • backend APIs and services
  • Python
  • FastAPI
  • data store technology
  • PostgreSQL
  • Elasticsearch
  • LLM-powered systems
  • GenAI concepts
  • RAG
  • vector stores
  • guardrails
  • evaluation frameworks
  • observability frameworks
  • AI/ML patterns safely within production services
  • AI/ML‑enabled solutions at scale

Other signals

  • building and operating shared generative AI platform
  • design and deliver backend services powering AI capabilities
  • RAG pipelines, vector stores, LLM integrations, and platform guardrails
  • production code, mentor junior engineers, and drive well-scoped projects