Description

Design and deploy production-grade LLM and agentic applications for content management workflows, owning the architecture from prototyping to cloud deployment.

This role is on-site.

Responsibilities

• Partner with product owners to scope ambiguous business ideas into testable AI requirements for content and media workflows.
• Prototype and deliver proofs-of-concept for use cases such as drafting, summarization, taxonomy mapping, and moderation.
• Design production architectures for LLM and agentic applications, including retrieval strategies, orchestration topologies, and human-in-the-loop checkpoints.
• Build and operate multi-agent systems using orchestration frameworks, ensuring robust tool calling, state management, and failure recovery.
• Deploy and manage AI solutions on cloud platforms with CI/CD, containerization, and cost governance for token- and GPU-intensive workloads.

Required Skills

• 4 to 6 years of experience in conversational AI, including at least 2 years specifically in content management systems.
• Proficiency in Python for AI/ML development and Spark for data processing.
• Experience deploying small language models (SLMs) via vLLM, Ollama, or Triton.
• Strong skills in prompt engineering, RAG, and parameter-efficient fine-tuning (LoRA, QLoRA).
• Hands-on experience with vector stores such as Azure AI Search, OpenSearch, Pinecone, or pgvector.
• Proficiency in CI/CD pipelines and containerization for Azure, AWS, or GCP.
• Familiarity with orchestration frameworks like LangGraph or CrewAI.
• Experience implementing guardrails, PII protection, and adversarial testing for secure AI deployments.

Preferred Skills

• Experience with evaluation harnesses, regression testing for model changes, and telemetry instrumentation.
• Knowledge of content enrichment, metadata management, and editorial workflow automation.
• Familiarity with asset localization, search discovery, and content moderation in media environments.

Key Skills

Education

Bachelor's degree