Staff Software Engineer, Data/ML

About Us

XOXO is a research lab building the interface of intelligence for everyday life. We're a stealth team of hardcore engineers, designers, and researchers discovering solutions to novel problems impacting life outside work.

With a recent breakthrough across infrastructure, architecture, and the model layer, we’re hiring serious builders to develop the interface and application layer that will bring our vision to life.

About the Role

We’re hiring an experienced Data/ML Engineer to build the data foundation and production-grade ML systems that make our product feel instantaneous, reliable, and smart. This is a full-cycle role: you’ll architect high-throughput pipelines, define durable data models and contracts, and deploy low-latency feature and model serving that holds up under real-world load.

You’ll work closely with the founders and early engineering team to take prototypes into production—turning messy real-world signals into trustworthy datasets and real-time capabilities that power core product experiences.

What You’ll Do

• Build and run high-throughput batch + streaming pipelines for analytics, training, and product signals.

• Own real-time feature pipelines and online feature serving for low-latency inference.

• Design and govern dimensional data models; handle schema evolution without breaking downstream consumers.

• Implement and optimize model serving infrastructure to hit strict latency and reliability SLOs.

• Define and enforce event schemas, telemetry standards, and data contracts across teams.

• Partner with engineering, product, and research to turn ambiguous product needs into measurable, durable systems.

Skills & Qualifications

Minimum qualifications:

• Shipped data and/or ML systems from prototype to production with strong reliability, scalability, and latency outcomes.

• Strong command of data modeling and managing schema evolution in event-driven/streaming environments.

• Deep experience with streaming and real-time processing (e.g., Flink, Spark), including correctness under late/out-of-order data.

• Built or operated real-time feature pipelines and/or feature stores for low-latency inference.

• Hands-on experience with modern data warehousing (e.g., Snowflake, BigQuery) and telemetry/event data standards.

• Able to translate product and business questions into durable datasets, metrics, and contracts.

Preferred qualifications — we encourage you to apply if you meet some but not all of these: