Remote Senior Data Scientist, Computer Vision Pipelines

Blue River Technology; Santa Clara (CA); Hybrid; USD 209,862 - 275,000; Full time

Job summary

Blue River Technology is seeking a highly skilled Data Scientist/ML Engineer to define and curate datasets for autonomy-focused computer vision, collaborating with data engineers, field staff, and software teams throughout the ML lifecycle.

You will design scalable data pipelines, analyze large image datasets, and deploy cloud-based solutions using modern tools like Docker, LabelBox, and open source models.

Qualifications

• Master’s degree in Math/Physics/Data Science or related field with 5 years of related experience.
• Implement and deploy computer vision and ML-based data pipeline systems using semantic segmentation, image & video classification, object detection, supervised, and unsupervised learning (5 yrs).
• Experience with data engineers, data scientists, software engineers, and field staff through the lifecycle of developing and deploying a machine learning system (4 yrs).
• Perform non-parametric statistical tests and analysis on large image-based data sets using sklearn, scikit-image, scipy, and OpenCV (3 yrs).
• Write technical documentation, tutorials, and summaries to train data collection teams and conduct on-site training (3 yrs).
• Deploy scalable cloud-based solutions to mine, preprocess, resize, crop, rectify, and filter image-based data sets (5 yrs).
• Implement code using Python libraries, including NumPy, SciPy, OpenCV, Pandas, Seaborn, Matplotlib, CUDA, Pytorch, and TensorFlow (5 yrs).
• Design, implement, debug, and deploy stereo image-based data pipelines using Apache TeamCity, AWS Airflow, Redis, Google appsheet, Data bricks datatables, Celery, and advanced search solutions on LabelBox with open source models such as CLIP and BLIP (6 mos).
• Design, build, and debug custom Python pipelines using Python Functools for processing large image datasets, deploy these pipelines using Docker and Docker-compose (1 yr).
• Use statistical sampling algorithms to design efficient data collection methods for large stereo camera-based image datasets and coordinate data collection (6 mos).
• 10% domestic travel required. Position is remote, but there is domestic travel to test/training sites required and regular in-office time (about once a week) to interact with local workstations and participate in in-person meetings.

Responsibilities

• Define, curate, and manage datasets of images, sensor data, and scenarios that are designed to increase the trust and safety of autonomy.
• Work closely with data engineers and field data capture technicians to mine fleet data and identify open needs.
• Define frameworks for cataloging and searching scenario-based data to serve multiple stakeholders, including computer vision and robotics teams.
• Monitor, investigate, and fix data ingestion issues related to dataset curation for training and testing computer vision algorithms.
• Investigate data quality and actively participate in conceptualizing and developing short and long-term solutions.
• Provide data and infrastructure support to internal teams.
• Provide guidance to improve the stability, security, efficiency, and scalability of image data pipelines.
• Improve code quality through writing unit tests, automation, and performing code reviews.
• Examine the correlation between customer experience and virtual performance in like scenarios; adjust as needed. Ensure that defined safety and productive test cases are adequately covered with curated scenarios.

Skills

Semantic segmentation

Image & video classification

Object detection

Python (NumPy/SciPy/OpenCV)

CUDA

PyTorch

TensorFlow

Data pipelines

Cloud deployment

Docker

LabelBox

Apache TeamCity

AWS Airflow

Redis

Education

Master’s degree in Math/Physics/Data Science

Tools

Apache TeamCity

AWS Airflow

Redis

Google AppSheet

Databricks

Docker

Docker-compose

LabelBox

Job description

Blue River Technology is seeking a highly skilled Data Scientist/ML Engineer to define and curate datasets for autonomy-focused computer vision, collaborating with data engineers, field staff, and software teams throughout the ML lifecycle.

You will design scalable data pipelines, analyze large image datasets, and deploy cloud-based solutions using modern tools like Docker, LabelBox, and open source models.