Machine Learning Engineer III, Metadata Extraction
Core
Design, build, and operate production LLM agents and machine learning systems to extract structured metadata from unstructured enterprise documents.
Role type
Senior IC machine learning engineer (metadata extraction)
Builds
AI-powered document understanding capabilities (OCR, LLM-based extraction agents) for Box's Intelligent Content Management platform
Domain
Enterprise content management, document intelligence, generative AI
Deliverable
production ML models
Required skills
Python, LLM application frameworks (LangChain, LangGraph), LLM APIs (Gemini, Claude, OpenAI), data preparation pipelines, prompt engineering, RAG, model evaluation, API development, cloud distributed systems, CI/CD, observability
Preferred skills
MLOps tooling, feature stores, workflow orchestration, experiment tracking, multi-provider gateways, semantic search, multimodal document processing, Kubernetes, Terraform, SQL, BigQuery, enterprise software experience
Technologies
LangChain, LangGraph, PyTorch, Pydantic, Pandas, NumPy, Gemini, Claude, OpenAI, GCP, AWS, Azure, Kubernetes, Terraform, BigQuery, Jira, GitHub, GIT
Responsibilities
Design, build, deploy, and maintain production LLM agents and ML services; develop data preparation pipelines and agentic workflows; evaluate and improve agent quality using offline/online metrics; monitor production systems for quality, latency, and drift; contribute to scalable service architecture; create technical designs and operational runbooks; participate in on-call rotation
Seniority
Senior, hands-on IC