Machine Learning Engineer - Computer Vision

CompanyCam

18h•Remote

About The Position

CompanyCam is seeking a Machine Learning Engineer with deep computer vision experience to join their ML team. Contractors use CompanyCam to capture millions of jobsite photos daily, and this role will involve transforming that visual data into structured understanding by building and deploying computer vision systems for image classification, document detection, segmentation, and multimodal embeddings. The engineer will own problems end-to-end, from data preprocessing and model training to evaluation and production deployment, making architectural decisions and seeing their work implemented in the product. Current and near-term challenges include segmentation, on-device model deployment, vision-language model integration, and building sustainable evaluation infrastructure.

Requirements

3+ years of experience shipping machine learning models to production (not just training them).
Experience with computer vision techniques including image classification, segmentation, and object detection.
Strong coding skills in Python with proficiency in PyTorch or TensorFlow and comfort with modern architectures (transformers, CNNs, etc.).
Strong SQL skills including joins, subqueries, window functions, and CTEs.
Proficiency in data analysis, cleaning, transformation, and feature engineering.
Experience with version control (Git), experiment tracking, and ML development best practices.
Ability to explain technical concepts to non-technical stakeholders through clear writing and presentations.
You live and work permanently in the U.S.

Nice To Haves

Embeddings, vector databases, and similarity search
On-device model deployment (e.g., Core ML, TensorFlow Lite)
MLFlow, Weights & Biases, or similar experiment tracking platforms
Amazon Bedrock or other cloud ML services
Ruby on Rails or JavaScript/React (for integration work)

Responsibilities

Design, train, and deploy computer vision models to production with well-understood performance, latency, and cost characteristics.
Own the full ML pipeline: data preprocessing, feature engineering, model selection, training, evaluation, and deployment into sustainable inference services.
Conduct discovery spikes to validate feasibility and inform go/no-go decisions before committing to full development.
Integrate ML solutions with observability tooling, establishing and maintaining benchmarks to measure improvement and compare approaches.
Build automated, self-sustaining ML pipelines. Models should train, evaluate, and deploy with minimal manual intervention.
Inform build-vs-buy decisions with both technical rigor and business context, understanding when in-house models create competitive advantage vs. when vendor APIs are sufficient.
Collaborate with software engineers, data engineers, and product stakeholders to integrate ML solutions into CompanyCam's platform.
Communicate clearly with non-technical audiences about feasibility, requirements, and trade-offs of proposed solutions.