The opportunity
At Instacart, we invite the world to share love through food because we believe everyone should have access to the food they love and more time to enjoy it together. Where others see a simple need for grocery delivery, we see exciting complexity and endless opportunity to serve the varied needs of our community.
What you'll do
Define Ads ontologies and canonical metrics across campaigns, budgets, bids,: creatives, audiences, and placements to power consistent reasoning and recommendations.
Build robust dbt models and curated marts in Snowflake (or BigQuery) with: clear data contracts, tests, SLOs, and monitoring; orchestrate pipelines with Airflow.
Ingest, structure, and enrich unstructured Ads content; publish: retrieval-ready datasets leveraging managed search/vector services to enable high-quality grounding.
Design and evaluate RAG workflows (hybrid search, re-ranking) with explicit: quality and latency targets; iterate via offline/online experiments to improve performance.
Design agent reasoning, tools, and policies for Ads use cases, including: human-in-the-loop approvals and guardrails that balance safety, cost, and speed.
Establish evaluation suites and dashboards tracking precision/recall,: calibration, hallucination rate, latency, and cost; drive continuous model and system improvements.
What they're looking for
- Run A/B and uplift experiments to quantify business impact; own KPIs such as: ROAS lift, pacing accuracy, RCA precision/recall, forecast MAPE, and time-to-insight.
- Partner with Ads Product, Ads Engineering, Sales leadership, Data: Engineering, and R&D to align roadmaps, de-risk launches, and ship high-quality production agents.
- Own your workstream end to end—from data and pipelines to the UI and agent: logic—delivering secure, observable, and reliable tooling that’s adopted by the field.
- –7 years of experience in analytics engineering, data science, or applied AI,: with advanced proficiency in Python and SQL.
