Data Engineer, On-chain Measurement

A scheduled sampler reads finalized blocks, attributes transactions to venue programs and counts what failed. The output is published as JSON and CSV, cited with its measurement date, and used in our own routing decisions. The work is to widen what it measures without ever letting a number go out that we cannot defend.

What you would own

• Extend sampling to new dimensions such as buy and sell pressure per venue, or depth distribution at launch
• Improve statistical honesty: sample sizing, confidence, and being explicit about what a figure does not support
• Maintain the public dataset and the archive so historical figures stay reproducible
• Write the limitations section for every new metric, before the metric ships

What we need

• Strong data engineering fundamentals and comfort working with raw blockchain data structures
• The instinct to check whether a sample is large enough before drawing a conclusion from it
• Python, Rust or PHP to a production standard

Nice to have

• Experience publishing an open dataset that other people depend on
• Background in measurement or statistics rather than only in pipelines