Key Responsibilities

Build and contribute to the AI context platform

• Implement end-to-end pipelines: ingestion → parsing/chunking → enrichment → embeddings → vector indexing → retrieval/serving

• Build andmaintainpatterns for incremental refresh, backfills, re-embeddings, deduplication, and lineage across unstructured sources

• Contribute to retrieval quality improvements (query strategies, hybrid search, metadata filtering) in partnership with AI engineers

Deliver semantic and governed data products

• Implement semantic layers (metrics/entities) that power BI and agent reasoning consistently

• Apply established data contracts and context contracts for AI inputs (schemas, metadata requirements, freshness, citation expectations)

• Ensure datasets and indexes are documented and reusable

Operational excellence

• Support reliability and performance across assigned workstreams: monitoring, alerting, runbooks, and incident response

• Contribute to cost and latency optimization across Snowflake and vector infrastructure

AI safety and compliance

• Apply security-by-design patterns: RBAC/ABAC, PII redaction, retention controls, and audit logging

• Follow established guardrails for AI access to enterprise knowledge in coordination with Security/Legal/Compliance

• Ability to travel as needed up to 4 times per year.

Required Qualifications 

• Bachelor's Degree in computer science, engineering, or related field of study

• 3–6 years in data engineering or data platform roles with strong hands-on delivery 

• Strong SQL and Python (or Scala/Java); solid production engineering habits 

• Hands-on experience with Snowflake, including pipeline design, data modeling, and operating at scale in a production environment 

• Experience designing and operating cloud data pipelines at scale 

• Experience working with unstructured data processing and search/retrieval concepts 

• Clear communicator who can work effectively across technical and functional teams 

Preferred Qualifications 

• Hands-on experience with vector search and embeddings (pgvector/Pinecone/Weaviate/OpenSearch/Elastic) and retrieval patterns (semantic retrieval, hybrid search, reranking) 

• Experience supporting LLM applications (RAG, agent tool interfaces, evaluation/observability) 

• Familiarity with knowledge graphs, semantic modeling, or metrics layers 

• Experience in regulated environments and data governance programs 

• Exposure to dbt, Iceberg, or other lakehouse/semantic layer tooling alongside Snowflake 

Example Success Measures 

• Measurable improvement in AI outcomes: higher retrieval precision/recall, better citation coverage, fewer "missing context" failures 

• Reduced latency/cost per retrieval and improved platform reliability (SLO attainment, lower MTTR)

• Consistent application of semantic definitions and context contracts across assigned workstreams

• Delivery quality: production-ready outputs with minimal rework, well-documented and maintainable

Behavioral Attributes 

• Eager to learn the domain: Proactively builds familiarity with healthcare processes, terminology, and KPIs — can engage credibly with SMEs and ask the right clarifying questions 

• Collaborative and stakeholder-aware: Works well with engineers, consultants, and functional partners; communicates progress and flags risks clearly 

• Consultative problem-solver: Approaches requests with a "diagnose before prescribe" mindset — proposes options and works toward durable solutions rather than one-off fixes 

• High ownership and follow-through: Treats reliability, documentation, and operational readiness as part of the work; finishes what they start; holds a high bar for production quality 

• Clear communicator: Can go deep with engineers and explain concepts plainly to non-technical partners; writes solid docs and runbooks 

• Pragmatic builder: Biases toward shipping value in iterations, validating with users, and improving based on feedback 

• Comfortable with ambiguity: Adapts quickly in evolving AI/data product environments and turns unclear goals into actionable tasks 

• Integrity and stewardship: Handles sensitive data responsibly and respects established governance patterns