Do you want to empower organizations to build smarter compensation strategies while ensuring fair pay for all employees?

Syndio is the leading pay governance and compensation intelligence platform. We help organizations make better pay decisions at every stage of the compensation lifecycle, from leveling and offers to promotions and merit. Our platform gives HR, compensation, and finance leaders the data and decision support they need to govern pay fairly, compliantly, and with confidence. We partner with many of the world’s most recognized and respected enterprises, helping them implement leading-edge compensation solutions with expert guidance and analyzing pay for over 10 million employees across the world.

Join us in our mission to help companies make smarter pay decisions they can trust!

About the Role

Essentials is our pay equity product, and its analyses are relied on for high-stakes decisions. Delivering that means running complex statistical work across datasets that routinely reach hundreds of thousands of records, fast enough that a customer can iterate on a real question instead of waiting overnight for a job to finish.

This role is for the engineer who makes that computation fast, scalable, and trustworthy. You'll work on the workflows that move large datasets through ingestion, validation, modeling, and reporting: making them performant, making them observable, making them resilient when a single customer's dataset is an order of magnitude larger than anyone planned for. The architecture is genuinely complex, and you'll be one of the people who makes it more tractable rather than less.

The second half of this role is agentic. We are investing heavily in AI agents across the company, and Essentials is one of the highest-leverage places to apply them. You'll build the tooling that lets our agents carry out Essentials workflows — the harnesses, the guardrails, the context, and the evaluation that make an agent's output something we can stand behind, building on the AI capabilities our AI team provides.

This is a Staff role, sitting one level below Principal, and it carries the scope and expectations that implies. Success is measured two ways: how much larger and more complex a dataset we can handle without degrading, and how much of the Essentials workflow our agents can carry reliably.

What You'll Own

High-Performance Data Workflows

Build and optimize workflows that run complex mathematical and statistical calculations over datasets of hundreds of thousands of records — and design them so the next order of magnitude doesn't require a rewrite.
Own performance end to end: query plans and indexing in Postgres, batching and parallelism, memory behavior, caching strategy with Redis, and the difference between an algorithm that's clever and one that's actually fast on real customer data.
Make correctness verifiable. These results carry real consequences for our customers — performance work that quietly changes an output is worse than no performance work at all.

Orchestration & Scale

Work with the technologies that make large workflows dependable: message and queueing systems, workflow orchestration, job scheduling, retries, idempotency, and backpressure.
Design for partial failure. Long-running multi-stage pipelines fail in the middle; the interesting engineering is in what happens next.
Build the observability that makes a slow or wrong run diagnosable — timing, lineage, and intermediate state you can actually inspect.

Agentic Tooling for Essentials

Build the tooling that lets our agents execute Essentials workflows: data validation and remediation, analysis setup, exception triage, report generation, and the long tail of work that currently requires a human to babysit it.
Design the harnesses that make agent output trustworthy in a domain where being confidently wrong is unacceptable — evaluation sets, verification steps, constrained tool surfaces, and clear human review points.
Give agents the context they need to be productive across a large codebase: documentation, schemas, conventions, and project state that stay current.