Senior Computer Vision Data Scientist

Overview

We are seeking a full-time Senior Computer Vision Data Scientist at Garmin's U.S. Headquarters in the Kansas City area. This candidate selected for this role, in the Algorithm Technology Group, will use camera data to develop algorithms for features used in Garmin products across multiple segments and product lines.

Essential Functions

• Serves in a leadership capacity as an individual contributor for carrying out software development in C++ and Python, using machine learning frameworks like PyTorch and Tensorflow
• Provides technical leadership and strategic project planning for data science initiatives within computer vision
• Researches problems and implement appropriate algorithmic solutions, balancing traditional and machine learning-based computer vision approaches
• Collaborates with colleagues on software/algorithm design approaches and implementations
• Ensures data quality, adherence to governance standards, and compliance with data privacy and AI-related regulations
• Evaluates various design tradeoffs including power consumption, hardware acceleration availability, cost, reliability, and time to market
• Provides reliable solutions to a wide range of difficult problems using sound problem solving techniques
• Coordinates closely with product team engineers to integrate software deliverables into projects
• Works with data annotation platforms and direct annotation efforts as needed
• Coordinates closely with project stakeholders to evaluate software deliverables
• Collaborates and participate in peer code reviews and technical assessments in areas of expertise, providing comments and suggestions to support software designs and new technologies
• Offers process improvement suggestions and author new procedures as appropriate
• Mentors and/or supervises less experienced Software Engineers (Software Engineer 1 or 2 and Interns)

Basic Qualifications

• Bachelor's Degree in Computer Science, Computer Engineering, Software Engineering or related field AND a minimum of 5 years relevant experience OR an equivalent combination of education and experience
• Mastered proficiency in writing software in C++ and Python
• Expertise in computer vision libraries such as OpenCV
• Expertise in machine learning frameworks such as PyTorch, Tensorflow, Keras, or ONNX.
• Experience evaluating algorithm and model performance with a variety of statistical techniques.
• Demonstrated capability to offer peer technical assessments in areas of expertise, new technologies and software designs
• Demonstrated strong and effective verbal, written, and interpersonal communication skills
• Must be positive, detail oriented, organized, team oriented and a driven problem solver, multi-tasker, and self-starter with the ability to prioritize and lead in a fast paced, deadline-driven environment
• Excellent academics (cumulative GPA greater than or equal to 3.0 as a general rule)

Desired Qualifications

• Possess a fundamental understanding of linear algebra, quaternions, and 3D geometry.
• Possess a fundamental understanding of the operation and nuances of image sensors, lens optics, and the camera calibration process.
• Experience generating intrinsic and extrinsic camera matrices and estimating camera position and pose.
• Experience optimizing algorithms for embedded systems and servers.
• Experience porting machine learning models to neural processing units and to cloud deployments.
• Experience generating synthetic datasets for model pre-training.
• Experience adapting algorithms for improved performance in adverse weather and lighting conditions.

Garmin International is an equal opportunity employer. Qualified applicants will receive consideration for employment without regard to race, religion, color, national origin, citizenship, sex, sexual orientation, gender identity, veteran's status, age or disability.

This position is eligible for Garmin's benefit program. Details can be found here: Garmin Benefits