Member of Technical Staff, Product Engineering

San Francisco Tensor Company•San Francisco, CA

About The Position

SF Tensor is hiring a Member of Technical Staff for Product Engineering to own the product experience for Model Foundry, its platform for managing machine-learning research runs. Model Foundry helps researchers move from an idea to a running experiment, including configuring and queuing jobs on suitable hardware, streaming logs, and versioning experiments for later reproducibility. The role owns the product end to end, from the interfaces researchers use to the services behind them, with an emphasis on making the software reliable, polished, and enjoyable for extended daily use. The product serves SF Tensor's customers, internal research team, and forward-deployed researchers. The work involves processing telemetry and logs from thousands of GPUs across NVIDIA, AMD, TPU, and Trainium, including runs lasting weeks or months, and making data accessible to both users and their agents. The team works primarily in person from its San Francisco office; relocation assistance is offered. This full-time position has a base salary range of $225,000–$275,000, plus equity and benefits.

Requirements

Proven experience shipping complete products; ideally, be able to share an example.
A strong eye for design and motion, including attention to animation, spacing, and visual quality.
Solid backend skills, including the ability to build APIs and troubleshoot production issues.
Enough infrastructure experience to work with AWS, containers, and Kubernetes.

Nice To Haves

Experience with Bun or eagerness to learn it.
A background in developer tools, infrastructure products, or products for technical users.
Exposure to machine-learning training workflows or tools such as Weights & Biases, Ray, or Slurm.
Comfort using Figma to create mockups when needed.

Responsibilities

Take full ownership of Model Foundry product areas, including scoping, designing, building, and iterating on them.
Build clean, fast, polished web interfaces with NextJS.
Design researcher-facing experiences such as run dashboards, live log streaming, experiment comparison, evaluation results, and cluster and storage views.
Build conversational and commit-driven entry points that turn a Slack message or commit into a training run.
Polish product details, including loading states, animations, and edge cases.
Work directly with forward-deployed researchers and the internal training team, incorporating their feedback.
Make product tradeoffs independently, balancing speed of delivery with additional polish.
Build APIs and troubleshoot production issues.
Work with telemetry, logs, and charts from large-scale GPU training runs and make the data accessible to users and their agents.