AI/ML Engineer

Project Overview:

Data/ML engineer embedded on a product team building customer intelligence and personalization platform for restaurants. You'll own the data layer — from raw transaction data to production ML models — powering merchant-facing features like customer segmentation, behavior insights, diner profiles, and personalized communications. Responsibilities:- Build and maintain customer data pipelines that unify transaction, reservation, loyalty, and marketing data into a clean, queryable layer              - Develop and operate the Customer Personalization Engine in Snowflake Cortex or similar technology — trend analysis, behavioral segmentation, and recommendation models                                                           - Build the analytical foundation for Actionable Customer Behavior Insights — surface patterns from transaction history that translate into merchant-facing recommendations                                                                            - Support Global Diner Unification by designing cross-merchant customer identity resolution approaches - Partner with full-stack engineers to define APIs and data contracts that power merchant-facing features (Pre-Shift Report, Diner Profiles, Segmentation Tool)     - Own data quality, observability, and pipeline reliability for customer data products                                         - Advise the team on what's feasible with available data vs. what requires new collection — be the person who says "we have this" or "we need to build that pipe first"    Competencies - Snowflake or other datawarehousing experience is required — warehousing, transformations, UDFs, and working within Snowflake's or similar ecosystem day-to-day - Hands-on experience with Snowflake Cortex or comparable ML-in-warehouse tooling (Databricks ML, BigQuery ML) - Experience with dbt or equivalent transformation frameworks — building, testing, and maintaining modular data models that serve analytics and ML features - Familiar with feature engineering for ML — transforming raw transactional data into meaningful signals (recency, frequency, monetary value, behavioral patterns) - Built production ML systems — not just notebooks. Has shipped models that other teams and products depend on. - Comfortable with data modeling for messy, multi-source customer identity problems (POS + reservations + loyalty + marketing = one customer) - Can work embedded on a product team, not just a data team — comfortable pairing with frontend/backend engineers, scoping what's realistic in a sprint, and shipping incrementally - Familiar with privacy considerations around customer data aggregation and cross-merchant data sharing - Strong communicator — can explain data constraints and tradeoffs to PMs and engineers without data backgrounds Nice to Have - Experience in restaurant/hospitality tech or POS systems - Experience with real-time or near-real-time feature serving (not just batch)