Connecting human data with machines through a new ingestion layer.
ML Infrastructure Engineer
Unstructured human data remains locked in formats that choke machine learning models. We need an engineer to forge the pipelines that translate this raw information into something PyTorch can actually digest. You will architect and scale the Kubernetes-backed backbone powering our ingestion systems, bridging the gap between chaotic inputs and reliable inference. This is fundamentally about building the rails that move real-world data into production ML.
What they're looking for
- 3+ years of experience building ML infrastructure or data pipelines in Python.
- Deep working knowledge of Kubernetes and Docker for orchestrating distributed workloads.
- Proven ability to scale PyTorch training and inference pipelines from development to production.
- Comfortable operating on-site in San Francisco within a tightly-knit, growth-stage team.