About the job
Join Pinterest’s IT Enterprise Systems team to build analytics engineering and GenAI capabilities that power our go-to-market and corporate operations. You’ll create trusted data models and pipelines across core enterprise platforms and layer in LLM-enabled workflows with clear evaluation, observability, and guardrails.
Responsibilities
Build production-grade GenAI-enabled analytics solutions that integrate enterprise data sources (e.g., Salesforce, Gong, Oracle, IBM Planning Analytics, and other SaaS platforms).
Establish end-to-end LLM pipelines (retrieval/orchestration) with evaluation frameworks, observability, and validation guardrails to ensure reliability and safety.
Lead technical scoping for GenAI use cases, assessing feasibility, accuracy expectations, risk, and ROI—then translating into clear technical plans.
Design and own analytics-ready data models, including dimensional modeling (star schemas, fact/dimension tables, SCD Type 2) that support reporting, forecasting, and downstream applications.
Develop and maintain robust data pipelines and orchestration (Airflow and/or dbt or similar), including data quality checks, SLAs, monitoring, and failure recovery.
Write and optimize complex SQL for transformation and analysis across enterprise datasets (joins, window functions, CTEs, performance tuning).
Partner closely with business stakeholders, turning ambiguous questions into well-scoped requirements, success metrics, and deliverables; communicate tradeoffs and constraints clearly.
Use AI tools to accelerate development responsibly (e.g., drafting SQL/Python, iterating on prompts, summarizing findings), while applying rigorous testing and verification.
Define best practices for GenAI in analytics engineering, including prompt patterns for analytics tasks, model selection, and cost/latency optimization.
Qualifications
Minimum
8+ years in analytics engineering, data engineering, data science, or related roles (or equivalent experience).
Bachelor’s degree in Computer Science, Statistics, Data Science, Mathematics, or a related field (or equivalent experience).
Strong understanding of LLM fundamentals and practical tradeoffs, including:
Capabilities, limitations and common failure modes.
Tokenization, context windows, latency/cost considerations.
RAG vs. fine-tuning vs. prompting/tools for use cases.
Expert SQL skills (complex joins, window functions, CTEs, query optimization).
Hands-on experience with dimensional modeling (star schemas, SCD Type 2) and workflow orchestration tools such as Airflow, dbt.
Demonstrated ability to validate AI-assisted output.
High integrity and ownership to handle sensitive enterprise data appropriately and remain accountable for final outputs.
Strong stakeholder communication with the ability to explain technical constraints to non-technical partners.
Experience with GenAI coding tools (e.g., Cursor, Claude Code) to accelerate development while maintaining code quality and maintainability.
Preferred
No preferred qualifications listed.