Senior Computer Vision Engineer (Embedded / Edge AI) — NVIDIA Jetson Orin

ViaTouch Media (VICKI) is an AI-powered IoT self-checkout company backed by one of the world’s largest payment providers. We’re hiring an immediate Senior Computer Vision Engineer to join our engineering team and help drive a major platform optimiztin our current Raspberry Pi-based CV stack/NVIDIA Jetson Orin device family (edge GPU acceleration, real-time inference, production reliability).

Must be authorized to work in the United States.

Company site: www.getvicki.com

Job Type: Full-time

Locations: New York / New Jersey (preferred), San Diego.What You’ll DoCore ownership: Raspberry Pi → NVIDIA Jetson Orin

• Lead and execute the migration of production computer vision pipelines from Raspberry Pi to NVIDIA Jetson Orin (JetPack-based Linux environment).
• Refactor/optimize inference services to leverage GPU acceleration and edge deployment best practices (throughput, latency, thermals, reliability).

Computer vision + machine learning (retail-first)

• Research, design, train, and ship object detection and object tracking models tailored for retail environments (SKU/product recognition, multi-item scenes, occlusions, glare/reflective packaging, variable lighting, crowded bins/shelves).
• Design and evolve datasets (collection strategy, labeling specs, QA, augmentation, hard-negative mining, active learning loops).
• Benchmark models across speed, robustness, and accuracy, and define acceptance criteria for production rollout.

Edge deployment + performance engineering

• Deploy and run optimized models on embedded hardware, including:
• Model conversion and deployment workflows (ONNX, TensorRT, FP16/INT8 optimization as applicable)
• Real-time video pipelines (multi-camera and high-FPS scenarios when needed)
• Troubleshoot production issues spanning model behavior, camera feeds, inference pipelines, and device-level constraints.

Leadership

• Lead/mentor computer vision engineers and collaborate cross-functionally with embedded, backend, product, and operations teams.
• Set engineering standards for code quality, experimentation hygiene, and production readiness.

Required Qualifications

• MS or PhD in Computer Science, Electrical Engineering, or related field with a strong focus in computer vision / deep learning / machine learning.
• 4+ years professional experience in Deep Learning / Computer Vision / Machine Learning with production responsibility.
• Retail computer vision experience (required): prior hands-on experience training object detection models on retail objects/products (e.g., consumer packaged goods, SKU-level product recognition, shelf/bin/basket scenes), including dataset design and model iteration based on real-world edge-case failures.
• Embedded/Edge experience: 1+ year professional experience deploying CV/ML to embedded devices (must include performance constraints and production troubleshooting).
• Strong ability to take models from training to production deployment, including evaluation and iteration loops.

Required Programming Languages (must-have)

• Python (3.x) — training, experimentation, tooling, and/or production services
• C++ (C++14/17) — performance-critical inference/pipeline code on-device
• CUDA / GPU programming fundamentals — ability to reason about GPU acceleration and performance bottlenecks on Jetson-class devices

*(You don’t have to be a CUDA kernel wizard, but you must be able to work effectively in a CUDA-enabled stack and optimize for GPU deployment.)*

• Bash / Linux shell — device-level debugging, profiling, and automation

Required Frameworks / Tooling Experience

• Deep learning frameworks: PyTorch and/or TensorFlow (PyTorch strongly preferred for modern CV training workflows)
• Computer vision: OpenCV
• Deployment/optimization: TensorRT, ONNX (conversion + runtime considerations)
• Comfortable working in a Linux environment (Ubuntu-based stacks common for Jetson)
• Version control and collaboration: Git

Preferred (Strong Pluses)

• Direct experience deploying on NVIDIA Jetson platforms (Orin / Xavier family), including JetPack ecosystem familiarity.
• DeepStream and/or GStreamer experience for real-time video analytics pipelines.
• Modern detector families and deployment patterns (YOLO-family, transformer-based detectors, lightweight/mobile architectures).
• Multi-object tracking implementations (e.g., DeepSORT/ByteTrack-style approaches) and practical tuning in real-world scenes.
• MLOps/experiment tracking and dataset/versioning discipline (e.g., W&B, MLflow, DVC—any equivalent).
• Experience building systems that must be robust in retail conditions: changing planograms, seasonal packaging changes, partial occlusions, motion blur, glare, etc.

Candidate Profile We’re Looking For

You’re a hands-on CV engineer who can train models, ship them, and make them fast and reliable on edge hardware—and you’re comfortable leading a team while driving a platform upgrade from CPU-constrained Raspberry Pi pipelines to GPU-accelerated NVIDIA Jetson Orin production inference.

If

Job Type: Full-time

Pay: $85,000.00 - $125,000.00 per year

Benefits:

• 401(k)
• Dental insurance
• Health insurance
• Paid time off
• RSU
• Vision insurance

Work Location: In person