Sr. Computer Vision Engineer
Core
Architect and train cutting-edge computer vision and vision-language models for retail object recognition, inventory tracking, and edge deployment.
Role type
Senior IC Computer Vision Engineer (Retail & Edge)
Builds
Production-grade object detection models, vision-language-action (VLA) pipelines, and optimized edge inference systems for retail applications.
Domain
Retail technology, Computer Vision, Multimodal AI
Deliverable
production ML models
Required skills
Custom YOLO architecture design and training, Open-source VLM fine-tuning (LLaVA, Qwen-VL, InternVL, PaliGemma), VLA pipeline construction, Edge model optimization (quantization, pruning), TensorRT/ONNX Runtime deployment, Production ML system engineering, Data pipeline and annotation workflow design, Technical mentorship
Preferred skills
NVIDIA Jetson deployment experience, Retail domain knowledge, PyTorch and modern training frameworks (Transformers, LitGPT, Unsloth), Efficient VLM inference (vLLM, llama.cpp, SGLang), Synthetic data generation, Model versioning and experiment tracking (MLflow, Weights & Biases), AWS cloud services (EC2, ECS, SageMaker)
Technologies
PyTorch, YOLO variants, LLaVA, Qwen-VL, InternVL, PaliGemma, TensorRT, ONNX, vLLM, Docker, Kubernetes, MLflow, Weights & Biases, AWS
Responsibilities
Design, train, and iterate on custom object detection models for retail environments; Fine-tune and deploy open-source vision-language models for product understanding and scene reasoning; Optimize state-of-the-art models for edge deployment via quantization and pruning; Build robust data pipelines and annotation workflows; Prototype new architectures and determine production readiness; Mentor engineers and drive technical decisions on CV infrastructure
Seniority
Senior, hands-on IC
