Senior Software Engineer - AI Core Engineering

Disney Disney · Media · Glendale, CA +4

Senior AI Engineer to build AI core capabilities and tooling that accelerate teams across Ad Technology. This role blends backend engineering, LLM orchestration, and developer enablement, focusing on creating shared agents, initializers, registries, and reusable services and web applications to make AI adoption fast, safe, and cost-effective. The work emphasizes common design principles, guardrails, governance, observability, and evaluation to enable teams to deliver high-quality AI solutions quickly.

What you'd actually do

  1. Define and implement shared design patterns for agents and services, including context management, retrieval, safety, and evaluation.
  2. Establish reusable blueprints that standardize how AI solutions are built across Ad Tech.
  3. Build production-ready agents, initializers, and libraries for common use cases (summarization, troubleshooting, semantic extraction).
  4. Develop high-quality APIs and SDKs that simplify adoption of AI capabilities by other engineering teams.
  5. Deliver core services for retrieval, memory, and orchestration, integrated with enterprise and consumer facing systems.

Skills

Required

  • Python
  • LLM APIs (e.g., OpenAI, Anthropic, Claude)
  • LangChain or LangGraph
  • AWS Bedrock or Azure AI Foundry
  • building scalable APIs or SDKs
  • developing reusable agents, libraries, or internal tooling
  • guardrails and evaluation techniques for LLMs

Nice to have

  • AI enablement tooling
  • observability for AI systems
  • AI gateway patterns
  • multi-agent orchestration
  • enterprise governance
  • developer enablement role

What the JD emphasized

  • track record of building production-grade systems
  • building scalable APIs or SDKs
  • developing reusable agents, libraries, or internal tooling adopted by multiple teams
  • Knowledge of guardrails and evaluation techniques for LLMs

Other signals

  • building shared agents, initializers, registries, and reusable services
  • standardize adoption, define patterns for multi-turn interactions
  • provide evaluation/traceability
  • embed guardrails, compliance checks, and prompt safety
  • implement model evaluation and observability frameworks