DeepWalk is a fast-growing venture that helps cities keep people safe using computer vision to map and monitor their sidewalks. They are seeking a Staff Software Engineer to lead the technical direction of their computer vision platform for automated sidewalk inspection, focusing on large-scale imagery, production ML systems, and data pipelines.
Responsibilities:
- Own the lifecycle of our computer vision models, including training, evaluation, deployment, and iteration
- Improve model performance in real-world conditions (noise, edge cases, data drift)
- Design and improve data pipelines that process thousands of miles of sidewalks and millions of images
- Lead architectural decisions for handling hundreds of terabytes of geospatial and visual data, including storage layout, pipeline reliability, and inference performance
- Set and evolve best practices around deployment, observability, system reliability, and scalability
- Act as a senior individual contributor and mentor, helping raise the technical bar across the engineering team
- Work closely with the CEO and operations team to turn business needs into clear technical priorities