6 hours ago
Boston, MA, USASenior / Staff+
Base Salary
$149k - $214k/yr
Responsibilities
- Explore data and workflows to identify opportunities for CV/ML-based solutions.
- Apply feature matching, geometric transformations, image registration, and other classical computer vision techniques to internal and customer-facing features.
- Prepare training data and train, fine-tune, and validate vision-based models.
- Combine VLMs, vision-based models, and classical computer vision techniques.
- Ship continuously monitored CV pipelines with evaluation harnesses and quality metrics.
Requirements
- 6+ years of experience building computer vision systems in Python using tools such as OpenCV, NumPy, and PyTorch or equivalents.
- Strong grounding in classical computer vision, including feature extraction and matching, geometric transformations, image registration, and robust estimation such as RANSAC.
- Strong grounding in machine learning and ML-based computer vision, including CNNs, transformers, and embedding spaces.
- Hands-on experience preparing datasets and training, fine-tuning, and evaluating vision-based models.
- Ability to work autonomously, move quickly, and take initiative.
- Strong verbal and written communication skills.
- Bachelor's or master's degree in Computer Science, Electrical Engineering, or a related field.
- Experience with scanned documents, OCR outputs, low-quality rasters, geospatial tooling, VLM/LLM API integration, or MLOps is preferred.
Benefits
- Hybrid Boston, MA work arrangement with at least three office days per week and remote-friendly flexibility.
- Company-funded medical, dental, and vision insurance, including no-cost medical plans.
- Health and dependent-care FSA.
- Unlimited PTO.
- 401(k) plan.
- Meaningful early-stage equity and a high-trust environment with substantial ownership.
Categories
About CivilGrid
CivilGrid builds a construction intelligence platform that aggregates utility, environmental, and geotechnical site data into a collaborative GIS-based system for utilities, civil engineers, and municipalities. Sold as enterprise software, it centralizes records from 3,000+ utilities, agencies, and cities with chain-of-custody documentation to accelerate due diligence and design. Headquartered in San Francisco, its customers include PG&E, Mark Thomas, and other engineering firms focused on infrastructure planning and delivery.