Senior Data Analyst - AI (Gen AI & Recommendation Systems)

We're looking for a Senior Data Analyst/ Analytics Engineer to own data and analytics across our Gen AI and Recommendation Systems work.

It's a hybrid role: you'll own the centralized reporting that turns data into decisions and build the pipelines and data models that feed it — defining the right metrics for each product we ship rather than waiting on others to prepare your data.

For Recommendation Systems, you'll bring enough ML understanding to engineer the right features and evaluation metrics, partnering closely with Data Scientists, ML Engineers, Product, and Backend teams.

Pipeline Architecture & Development: Build and maintain scalable, fault-tolerant batch and streaming pipelines that serve analytical and ML use cases.

Centralized Reporting & Metrics: Define the key metrics for each product we ship and build rock-solid centralized reporting around them, surfacing the trends and insights that matter.

Data Modeling: Design and own multi-layer data models (staging to feature-ready marts) that stay consistent and performant across ML models, dashboards, and APIs, handling schema changes cleanly.

Feature Store & ML Data Flows: Engineer the data flows that populate and update our ML Feature Store (and graph data where relevant) with the availability and low latency recommendation models need.

Experimentation & A/B Testing: Build the pipelines and metrics frameworks behind A/B testing — experiment schemas, assignment logging, and reliable metric computation for statistically sound results.

ClickHouse Mastery: Own ClickHouse as the domain expert — schema design, performance tuning, and fast queries for experiment aggregation and feature serving.

Streaming & CDC: Implement Change Data Capture (CDC) and event-driven flows (e.

g.

Apache Kafka) to keep data fresh where reporting and recommendations need it.

Orchestration & Automation: Build and manage workflows with modern orchestration tools (e.

g.

Mage AI, Airflow, Prefect) for reliable delivery and dependency management.

ML-Aware Support: Define and interpret the right offline and online ranking metrics, and engineer the features the models actually need.

Cross-Functional Collaboration: Partner with Data Scientists, ML Engineers, Product, and Backend to turn data requirements into production pipelines and actionable ML features.

4+ years as a Data/Analytics Engineer building data systems for analytics and ML.

Programming: Expert Python and advanced SQL.

BI & Visualization: Strong BI/visualization skills (e.

g.

Looker, Tableau) and good intuition for which metrics matter and how to present them.

Pipelines & Orchestration: Hands-on building pipelines with modern orchestration (Mage AI, Airflow, Prefect) — you build your own data, not just consume it.

Data Warehouse / ClickHouse: Deep production experience with ClickHouse (or BigQuery, Snowflake, or similar).

Data Modeling: Hands-on multi-layer modeling (raw, staging, marts) using Kimball, Data Vault, or OBT patterns.

Experimentation & A/B Testing: Solid grasp of experimentation frameworks — assignment, holdouts, metric pipelines, variance reduction.

ML Exposure: Good grasp of the ML lifecycle — how models consume data, how Feature Stores work (e.

g.

Feast, Hopsworks), and how to engineer features at scale, plus enough ranking-metric knowledge to support Recommendation Systems.

Nice to have: DBT for modeling and transformation.

Building or integrating A/B platforms (e.

g.

Statsig, Optimizely, GrowthBook, or custom).

Apache Kafka and CDC tools (e.

g.

Debezium, Maxwell).

Graph Databases (e.

g.

Dgraph, Neo4j, Amazon Neptune) and structuring data for them.

JavaScript or Go.