Senior Software Engineer, Biztech

Airbnb Airbnb · Consumer · United States · Software Engineering

Senior Software Engineer to build AI-powered employee experience platforms, focusing on democratizing AI for non-technical users. The role involves leading the design and implementation of LLM-powered features, including RAG pipelines and agent orchestration, and taking AI prototypes from concept to production at scale. Responsibilities include architecting and building production-ready AI/ML systems, developing agentic AI capabilities, implementing evaluation pipelines, and owning production AI systems end-to-end.

What you'd actually do

  1. Lead the technical design and implementation of LLM-powered features for OneChat and enterprise AI tools, including RAG pipelines, agent orchestration, and prompt optimization
  2. Partner with product managers, designers, and cross-functional teams to translate user problems into AI-powered solutions that serve Airbnb's global workforce
  3. Architect and build production-ready AI/ML-integrated systems, ensuring scalability, reliability, and low latency across multi-cloud environments
  4. Develop and iterate on agentic AI capabilities, including multi-step reasoning, tool use, and context-aware decision-making
  5. Implement evaluation pipelines and quality systems to measure model performance, safety, and user satisfaction

Skills

Required

  • Python
  • TypeScript
  • Go
  • Java
  • LLMs
  • fine-tuning
  • prompt engineering
  • embeddings
  • retrieval-augmented generation (RAG)
  • building and deploying production ML systems at scale
  • high availability
  • low latency requirements
  • backend and distributed systems expertise
  • API design (REST, GraphQL)
  • cloud infrastructure (AWS, GCP)
  • shipping AI-powered products from prototype to production
  • collaborate cross-functionally
  • influence without authority
  • communication skills
  • distill complex technical concepts

Nice to have

  • AI agent frameworks (LangChain, LangGraph, or similar)
  • agentic development patterns
  • foundation model APIs (OpenAI, Anthropic/Claude, Google)
  • ML evaluation systems
  • LLM-as-a-judge approaches
  • containerization
  • orchestration (Kubernetes)
  • infrastructure-as-code (Terraform)
  • enterprise-grade internal tools
  • developer productivity platforms
  • frontend technologies (React, Next.js)
  • full-stack AI product development
  • Publications at top AI/ML venues

What the JD emphasized

  • building production AI/ML systems
  • Track record of shipping AI-powered products from prototype to production

Other signals

  • building AI-powered employee experience platforms
  • democratize AI
  • drive innovation by taking AI prototypes from concept to production at scale
  • architect and build production-ready AI/ML-integrated systems
  • develop and iterate on agentic AI capabilities
  • own production AI systems end-to-end