Software Engineer, Evaluation Platform (Full Stack, UI)

Summary

Siri Agentic Evaluation Platform. Evaluation is no longer just a static test suite—it is a highly dynamic, massive-scale distributed problem. Our platform enables teams to run high-throughput agentic simulations, orchestrate multi-model judging pipelines, and generate real-time observability dashboards across billions of tokens and complex data types.

Description

We are looking for a Full-Stack Software Engineer to own the user-facing UI of our evaluation platform. Teams across Siri and Apple Intelligence use our web console to create, run, browse, and compare evaluation jobs at scale. You will set the front-end's engineering standards and help drive the service APIs and data queries behind agent flows, the UI, and other clients, working with the services and data teams.

The hardest problems here are data-dense. Scenario browsing, job comparison, and cost reporting are limited by how evaluation data is stored and queried, not by the UI layer. You will help shape the query and aggregation side with the data backend, then build the interface on top. You will partner with the pipeline, device, evaluator, and CLI teams rather than work inside their internals.

Minimum Qualifications

5+ years building production web applications. Strong TypeScript and a modern framework (React).

Experience with data-heavy UIs: large tables, filtering, comparison, and visualization.

Working proficiency in Python, with experience collaborating on typed API contracts (OpenAPI or generated clients) with backend teams.

Experience owning or leading a codebase or major area: code review, setting patterns, and mentoring contributors.

Strong cross-team communication.

BS or MS in Computer Science or equivalent experience.

Preferred Qualifications

Experience with analytical or relational data backends (for example, PostgreSQL and query or aggregation design). You should be able to reason about why a view is slow and fix it at the query layer.

Experience building internal developer platforms or analytics consoles.

Experience leading framework upgrades and migrations in a live codebase.

Familiarity with ML/LLM evaluation workflows, such as judging, traces, and cost or token metering.