Context Trainer (developer)

Takeda

Bengaluru, India
Embeddings/retrieval expertise
Content transformation
Metadata extraction
Curate and govern the context layer to improve answer quality and minimize hallucinations, while protecting data/PII

Job Summary

  • Curate and govern the context layer to improve answer quality and minimize hallucinations, while protecting data/PII.
  • Extract and curate content from enterprise sources using APIs and automation, defining chunking and metadata schemas, labeling guidelines, and golden Q&A sets.
  • Enforce data minimization, retention, and access controls, maintaining lineage and approvals per Responsible AI.

Matching Summary

Curate and govern the context layer to improve answer quality and minimize hallucinations, while protecting data/PII.

Skills & Requirements

Must-have

  • embeddings/retrieval expertise
  • content transformation
  • metadata extraction
  • labeling workflows
  • vector stores
  • REST APIs
  • Python coding proficiency

Nice-to-have

  • designing golden datasets
  • evaluation pipelines
  • software development lifecycle
  • technical documentation patterns

Key Requirements

  • 3+ years data/ML experience
  • strong documentation skills
  • runbook skills
  • Familiarity with privacy principles
  • Familiarity with data governance principles
  • Hands-on experience with labeling tools
  • data extraction from enterprise systems

Work Rights

Not specified

Tailored Resume

Cover Letter