Software Engineer, Data Visualization
Build, San Francisco
The role
We’re hiring someone to make it incredibly easy to see the dataset, collection, quality, and evals. You will build the analysis and visualization layer — from capture through slices, examples, and live pipeline health — so people look at the data instead of dumping to a notebook. If someone has to wait on you for a plot, the product is not done.
What you will do
- Design and ship a visualization suite: data collection, processing, analysis, and UI (web and/or CLI) for the dataset and the data engine
- Make looking at examples, batches, coverage, quality, and failures fast enough that research and ops actually do it
- Build views for collection health, taxonomy coverage, and eval results, wired to live pipelines rather than one-off dashboards
- Work with Head of Dataset & Quality, Evals Lead, research, and ops on what must be visible to make a decision