Software Engineer II

Warner Bros Discovery Warner Bros Discovery · Media · Atlanta, GA +1 · Technology

Software Engineer II on the AI Systems team at CNN (Warner Bros. Discovery) responsible for designing, building, and operating AI-powered applications and the platform capabilities that support them. This includes LLM-based content understanding, consumer-facing AI features, AI platform maturation (prompt management, evaluation, caching, versioning, deployment), and internal AI tooling. The role involves integrating ML models into production applications with guardrails, monitoring, and evaluation, and driving operational excellence for highly available, low-latency systems.

What you'd actually do

  1. Integrate ML models and foundation models into production applications with appropriate guardrails, monitoring, and evaluation
  2. Build the in-house, LLM-based content understanding capability that replaces a vended classification product — improving quality, expanding adoption to new business domains across CNN, and powering downstream use cases including ad targeting, content routing, and brand safety.
  3. Ship and iterate on AI-powered features that reach millions of users — article summaries, weather summaries, and the next set of audience-facing experiences — with the accuracy and trustworthiness CNN's audiences expect.
  4. Grow CNN's AI platform capabilities — prompt management, evaluation, caching, versioning, deployment, and tenant management — so that AI features across CNN can be built, tested, and operated reliably and efficiently.
  5. Build AI-assisted workflows for editorial, product, design, engineering, and business teams that drive measurable productivity gains.

Skills

Required

  • Python
  • backend engineering experience with data-intensive applications at web scale
  • relational databases (Postgres or equivalent)
  • NoSQL databases (DynamoDB or equivalent)
  • infrastructure as code (Terraform or equivalent)
  • design, build, and ship highly available, low-latency systems

Nice to have

  • integrating LLMs or foundation models into production applications
  • prompt engineering
  • evaluation frameworks
  • LLM observability
  • classification
  • content understanding
  • information retrieval systems
  • experimentation frameworks and A/B testing methodologies
  • machine learning or AI team experience
  • media, publishing, or news organizations background

What the JD emphasized

  • production-quality code
  • highly available, low-latency, and efficient software
  • integrating LLMs or foundation models into production applications

Other signals

  • building AI-powered applications and systems
  • shipping features that improve user experience and internal productivity
  • integrating ML models and foundation models into production applications