Context Trainer (Developer)
Core
Curate and govern the context layer (RAG/KBs, embeddings, metadata, labeling) to improve answer quality and minimize hallucinations while protecting data/PII.
Role type
Context Trainer (Developer)
Builds
RAG systems, knowledge bases, and evaluation datasets for enterprise AI applications
Domain
Enterprise AI, Retrieval Augmented Generation (RAG), Data Governance
Deliverable
production ML models
Required skills
embeddings and retrieval expertise, content transformation, metadata extraction, labeling workflows, vector stores, REST APIs, Python coding, data governance principles
Preferred skills
golden dataset design, evaluation pipelines, AWS Bedrock Knowledge Bases, software development lifecycle patterns
Technologies
OpenSearch, pgvector, Kendra, Chroma, Confluence, Jira, SharePoint, ServiceNow, qTest, AWS Bedrock
Responsibilities
Extract and curate content from enterprise sources using APIs and automation; Define chunking and metadata schemas; Run A/B experiments across vector stores; Enforce data minimization, retention, and access controls
Seniority
Mid-level, hands-on IC
