Principal Software Engineer, Data Products

Shippo is building logistics technology and infrastructure that connects merchants to carriers through a single API and dashboard. The Principal Software Engineer, Data Products will set engineering standards and design data-intensive systems integrating application services with data science, data engineering, AI, and machine learning systems. The role also involves turning analytical and prototype work into secure, reliable production systems and mentoring technical leaders.

Responsibilities:

• Set the bar in design review and code review by demonstrating it
• Raise the standard for testing, observability, and operational readiness across the Data org
• Turn repeated solutions into reusable primitives. Document them so the next team does not need you
• Grow the next set of technical leaders. Pair, mentor, and hand off ownership
• Design and build the interfaces where Shippo's application services meet our ML and data systems. Write the reference implementation, not only the design doc
• Author technical RFCs to guide data, API, and application architecture. Run the reviews that make those decisions hold
• Set the performance, security and reliability standards for how application services call data and ML systems
• Ensure analytical and data engineering work turn into systems that meet our bar for security, quality, maintainability, and performance
• Pragmatic approach working directly with data scientists and data engineers to find the friction in their workflow, then remove it
• Ensure our data products deliver customer and business impact. Balance speed to market against architecture, and say which one wins and why
• Advise on the development workflow from experimentation to production: train, track, register, validate, deploy, monitor, retrain
• Experience building agentic workflows. Prove them in production against real traffic with real evaluation. Harden the ones that deliver value and kill the ones that do not

Requirements:

• 10+ years of software engineering, with deep ownership of large-scale production systems. You have set the engineering standard on a team, not just met it
• Track record as a principal-level individual contributor. You set technical direction across teams, wrote the RFCs, ran the design reviews, and delivered multi-quarter roadmaps without direct authority
• Depth in distributed, data-intensive production systems. You have made design decisions driven by latency, throughput, availability, and cost, and you have experience operating them in production
• Deep Kubernetes expertise: experience building and deploying containerized production services, with the ability to navigate autoscaling, resource constraints, and resilience patterns to define deployment standards for ML workloads
• Strong working experience alongside data scientists and data engineers. You have taken analytical or prototype work and made it production grade
• Proven adoption and mentorship. Teams chose the standards and tooling you built because they were better, not because someone mandated them, and engineers got better from working with you
• Solid understanding of what a good ML development workflow looks like, enough to advise a team building one
• Experience building agentic AI solutions for customers, including evaluation harnesses, guardrails, and observability
• Hands-on Databricks, specifically with ML tooling
• Feature store patterns and online-offline consistency