Requirements
- 5+ years into your career, with models you trained running in production
- Own the ML pipeline end to end: data, annotation, training, evaluation, deployment, monitoring
- Train and ship detection and segmentation models in PyTorch
- Decide what data to collect and label; keep datasets and annotation tooling trustworthy
- Build evaluation that predicts real lab performance and make ship/don't-ship calls
- Find production model failures and feed them back into training
Benefits
- Generous PTO
Skills
Full job description
Compensation: $175k – $190k • 0.0% – 0.5%
About VistaPath
VistaPath is reinventing pathology by bringing augmentation and computer vision into the lab - saving customers time and money while creating a safer experience for patients. Our first product is a software-driven, hardware-enabled smart workspace that seamlessly interacts with technicians to revolutionize lab operations. We are looking for a pragmatic Machine Learning Engineer to own our computer vision pipeline end to end.
You Will:
- Own the ML pipeline end to end: data, annotation, training, evaluation, deployment, and monitoring.
- Train and ship detection and segmentation models (PyTorch) that find and measure tissue in the lab.
- Decide what data to collect and label, and keep our datasets and annotation tooling trustworthy.
- Build evaluation that predicts performance in the lab and make the ship / don't ship call.
- Find where production models fail and feed those failures back into training.
- Execute responsibilities in compliance with our Quality System and our Information Security System.
You Are:
- 5+ years into your career, with models you trained running in production.
- Skeptical of a good metric until you have looked at the failures.
- A generalist who writes the tooling and infrastructure your models need.
- A self-starter with a bias toward shipping.
- Comfortable making calls under ambiguity and equally willing to pause and ask.
- A direct communicator who flags risk early.
- Someone with strong opinions, weakly held.
Bonus Points:
- Experience deploying models to edge devices (TensorRT, ONNX).
- Experience with data versioning and annotation tooling (DVC, COCO).
- Experience bringing a medtech, healthtech, or digital health product to market.
We Offer:
- Ownership of ML at the company: your decisions ship to labs.
- Passionate and fun teammates.
- Flexible vacation and work schedule.
- $170-195k and equity.
To achieve our mission of radically improving the safety and efficiency of pathology, we need our team to be representative of the world. We welcome your unique perspective and experiences in shaping this product. We encourage people from all kinds of backgrounds to apply, including and especially candidates from underrepresented and non-traditional backgrounds.