Machine Learning Engineer, Computer Vision

Newxel

. Own and improve the video understanding pipeline, including object detection, multi-object tracking, player re-identification, jersey-number recognition, OCR, event and action recognition, temporal localization, shot and scene-boundary detection, replay detection, and camera-view classification .

Posted 10/22/2026full-timeWarsaw • PolandMid-LevelSeniorWebsite

Core Competencies

Role fit

Core Competencies

Use this summary to align your resume positioning with the role.

Demonstrates expertise in building and optimizing video understanding pipelines, with a strong focus on object detection, multi-object tracking, and model evaluation metrics. Proficient in deploying production ML systems using Python, PyTorch, and GPU optimization techniques.

Highest-signal resume keywords

Production ML Systems ExperienceComputer Vision ExpertisePython and PyTorch ProficiencyGPU Inference OptimizationVideo Processing at Scale

ATS Keywords

Tailor your resume

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills

Object DetectionMulti-Object TrackingPlayer Re-IdentificationJersey-Number RecognitionOCREvent RecognitionTemporal LocalizationModel Evaluation MetricsProduction DeploymentVideo Action Recognition

Soft Skills

Strong Written Communication

Tools & Technologies

TensorRTONNX RuntimeDockerFFmpegCloud GPU Infrastructure

Industry Keywords

Video Understanding PipelineModel Bake-OffsRegression SuitesA/B TestingSynthetic Data Generation

Tech Stack

Tools & technologies

AzureCloudDockerFFmpegPythonPyTorch

About the role

Key responsibilities & impact

Own and improve the video understanding pipeline, including object detection, multi-object tracking, player re-identification, jersey-number recognition, OCR, event and action recognition, temporal localization, shot and scene-boundary detection, replay detection, and camera-view classification
Optimize GPU inference for throughput and cost through batching, precision tuning, and compilation
Monitor production quality and latency and investigate failures and regressions
Build and maintain labeled sports benchmarks split by sport, league, and content type
Define and track metrics including mAP, F1, precision, recall, IoU, latency, and GPU cost
Run systematic model bake-offs and reproduce published results on company data
Write decision memos with recommendations, trade-offs, and licensing considerations
Own regression suites and shadow or A/B testing pipelines, validating model and configuration changes before deployment
Track new papers, model releases, and leaderboards and identify developments for evaluation

Requirements

What you’ll need

3–5+ years of experience shipping production ML systems, including 2+ years in computer vision, ideally working with video
Strong Python and PyTorch skills
Hands-on experience profiling and optimizing GPU inference using TensorRT and ONNX Runtime
Hands-on experience in at least two of: object detection and segmentation, OCR and scene-text recognition, or video action recognition and temporal localization
Experience building test sets, calculating mAP, F1, and temporal IoU, and running rigorous model comparisons
Ability to critically evaluate research papers and model cards and reproduce published results
Production deployment experience with Docker, queue-driven services, and cloud GPU infrastructure
Experience with video processing at scale, including FFmpeg, video decoding pipelines, and frame sampling
Strong written communication skills and ability to document technical trade-offs and make clear recommendations
Nice to have: Azure Service Bus and Blob Storage, message-driven architectures, vision-language models, multimodal understanding, domain-specific vision-model fine-tuning, self-supervised learning, synthetic data generation, CVAT, Label Studio, shadow deployments, A/B testing, and commercial model licensing

Benefits

Comp & perks

Direct, unconstrained ownership of the full ML lifecycle—from paper replication and benchmark creation to live GPU optimization
Production impact at scale, where model optimization lowers infrastructure costs and improves latency
High-rigor, evidence-driven engineering environment prioritizing A/B testing, shadow paths, and statistical validation