Xceleration is replacing a decade of legacy reporting with a modern data platform, and we need someone to own the build. Xceleration runs RewardStation, an employee recognition and incentive platform serving enterprise clients across 90+ countries and 30+ languages. Our reporting sits in two worlds right now: a 400+ report SSRS estate built on assumptions that no longer hold, and Maslow Insights, our new analytics product running on MotherDuck and SQLMesh. Your job is to run the second one and retire the first.
This is a hands-on role. You will write code every day. You'll also set the architecture, manage the roadmap, and sit across from internal stakeholders who depend on numbers being right. Everyone in leadership here builds. This is how we operate at our size. Headcount to grow a team is planned, and you'll hire it!
What you'll own
Maslow Insights - day-to-day platform strategy and operations. Our multi-tenant analytics product. You own pipeline reliability, data quality, tenant isolation, and the roadmap for what it becomes next. The modern data warehouse. Today: manifest-driven extracts from RewardStation production land as Parquet in S3, feed MotherDuck, and get modeled in SQLMesh. You'll harden that path, extend it, and make decisions about where it goes. The SSRS sunset. A 6-8 month program in 2027 to fold legacy reporting into Maslow Insights. The estate is 400+ reports, but most are unused or near-duplicates of a parent. This is a rationalization, not a port. Success looks like a small, defensible set of reports that people actually use and the discipline to retire the rest.
What we're looking for
We screen on skills, not years. If you've done the work, the résumé length doesn't matter.
Required
• You've built and operated production data pipelines end to end, and been on the hook when they broke
• Deep SQL and Python
• You work with AI tooling every day like Claude Code or equivalent.
• Experience managing stakeholders who don't speak SQL
• Willingness to delete things. Migration projects fail when someone ports 400 reports instead of asking which 40 matter
Strongly preferred
• DuckDB or MotherDuck in production
• SQLMesh or dbt - SQLMesh is our transformation layer
• SSRS, SQL Server, or comparable legacy BI migration experience
• Building data products that agents consume - MCP servers, semantic layers, LLM-facing interfaces
• Multi-tenant data platforms with real isolation requirements
What this role is not
This is not a data science or ML role. We're not hiring for model development. If your background is training models rather than building the platform underneath them, this isn't the right fit.
How we work
We are a small team. We directly communicate with one another, no layers to route around. Decisions get made in the room. We're aggressive about AI leverage - not as a talking point, but because a team this size can't ship what we're shipping otherwise. You'll spend most of your time with internal partners across product, operations, and client services. Occasional direct client conversations, but account teams own those relationships. We're a team of bright, driven professionals, each of whom brings unique perspectives, experience, ideas, and skills to our roles. Our company culture means everything to us - we call it CARE (Collaborative. Adaptable. Respectful. Efficient.) Every member of our team works hard to exemplify these qualities in every aspect of the work we do each day.
