Applied AI Engineer

Everseen•Herndon, VA

9d•Hybrid

About The Position

Everseen is a leader in vision AI solutions for the world’s leading retailers. The Applied AI Engineer will report to the VP of Engineering and help build the next generation platform that powers Everseen’s products. This is hands-on work at the core of what is shipped, involving designing, training, and deploying models into production that run against live retail video at massive scale. The role will utilize the latest advances in vision and multimodal AI, including vision language models and world models, and the systems that enable real-time operation. This is a high-impact role offering opportunities to tackle complex problems and see work deployed in thousands of stores, impacting customer journeys in real time. The position allows for career advancement by working alongside industry experts and scaling a successful global organization.

Requirements

3-5 years building and shipping machine learning or computer vision systems, with real time spent taking models into production rather than only doing research.
Working knowledge of vision language models (VLMs) and world models, and a clear sense of where they help and where they do not.
Strong programming skills in Python and hands-on experience with a modern deep learning framework such as PyTorch.
Solid grounding in computer vision and deep learning fundamentals, including training, evaluation and debugging of real models on messy data.
Deep understanding of computer vision and deep learning fundamentals including tuning, training, evaluation and debugging of real models on messy data.
Familiarity with model optimization and deployment tooling such as TensorRT, ONNX, CUDA and GPU inference.
Comfortable working with large scale video or image data and the infrastructure around it.
Solid understanding of software engineering principles, version control (Git), and CI/CD pipelines. Ability to integrate and maintain strict standards of code and model quality for easy long-term maintenance.
A results-first mindset. You care whether the model works in the store, not just on the benchmark.
Excellent communication skills coupled with the ability and desire to work independently in a fast moving, delivery focused team. You enjoy solving complex problems creatively.
Bachelor’s or Master’s degree in Computer Science, Machine Learning, Engineering or a related field, or equivalent practical experience.

Nice To Haves

Experience building real-time vision AI applications is a strong advantage.

Responsibilities

Design, train, evaluate and own vision and multimodal models for Everseen’s next generation platform, and take them from prototype through to production.
Apply vision language models and world models to real retail problems, and understand what holds up at scale and what does not.
Design and implement targeted solutions to optimize system efficiency, real-time performance, and output.
Build and optimize real-time vision AI pipelines that process live video across thousands of checkouts and stores.
Partner with research and platform engineers to move new model architectures into production under tight latency and cost constraints.
Run experiments, measure results honestly, and let the data decide what ships.
Optimize models for GPU and edge deployment using tools such as TensorRT and ONNX, trading off accuracy against speed and footprint.
Contribute to and collaborate on the shared codebase, review other engineers’ work, and help raise the bar across the team.
Keep up with new research in computer vision and multimodal AI, expand on experimental results presented by others and successfully transition from research to production.