Job Description
Experience: 12+ years
Design and operate ingestion pipelines that parse, chunk, embed, and index structured and unstructured content (text, PDFs, images, audio transcripts) into vector and hybrid search stores. Ability to design indexes, configure rerankers and full knowledge on vector search algorithms.
Own RAG data quality: freshness SLAs, deduplication, metadata and access-control tagging, embedding versioning, and re-index strategies with minimal downtime.
Build AIOps telemetry pipelines capturing prompts, retrieval results, latency, token cost, and user feedback to power drift detection, anomaly alerts, and cost dashboards.
Partner with Applied AI engineers to evaluate retrieval quality (recall@k, MRR, groundedness) and tune chunking, reranking, and query-rewrite strategies.
Enforce data governance: PII redaction, lineage tracking, retention policies, and compliance with client security requirements.
Knowledge graphs: Understand and have implemented knowledge graphs and ontologies. Cross data hyper graph structures and querying
Connectors: Understand modern data integration architectures, MCP servers, RESTful apis, GraphQL, streaming
Apache Airflow, Snowflake, databricks, Spark and similar tech stack implementation experience that include data observability tools like DataDog, AccelData, MonteCarlo etc