INETUM
Somos un Consultora digital, internacional y ágil. En la era de la post-transformación digital, nos esforzamos para que cada uno de nuestros 28.000 profesionales pueda renovarse continuamente.
Cada uno de ellos puede diseñar su itinerario profesional de acuerdo a sus preferencias, emprender junto a sus clientes para construir en la práctica un mundo más positivo, innovar en cada uno de los 27 países y conciliar su carrera profesional con su bienestar personal.
Nuestros 28.000 atletas digitales están orgullosos de haberse certificado Top Employer Europe 2024
Descripción del empleoSenior engineer with full, end-to-end technical ownership of the vision core of an industrial visual inspection product built on the NVIDIA platform. Combines deep expertise in unsupervised anomaly detection and defect segmentation with production-grade GPU inference optimization and complete model lifecycle management. Also able to design and lead the evolution of the pipeline orchestration toward a high-performance native C++ core, delivering real-time decisions on the factory floor.
RequisitosDeep Learning & Computer Vision
- Expert-level PyTorch: CNN / transformer vision architectures, training, evaluation and rigorous ONNX export (zero train/serve skew).
- One-class / unsupervised anomaly detection: PatchCore, EfficientAD, PaDiM, student–teacher, normalizing flows, and their real failure modes (reference-set contamination, threshold calibration with few or no defective samples, synthetic defects via cut-paste / DRAEM, over-rejection, drift).
- Supervised defect segmentation and detection (encoder–decoder, DETR-family): training, acceptance criteria and imbalanced datasets.
- Methodological rigor: AUROC / AUPRO alongside plant-level metrics (escape rate, false-reject rate) and regression validation against recorded data.
C++ & Real-Time Systems
- Expert-level modern C++ (C++17/20) for real-time vision pipelines, in addition to expert Python.
- High-performance systems design: native pipeline/orchestrator coordinating capture, pre-processing, inference and post-processing while keeping data in memory and avoiding unnecessary copies and hops.
- Hard latency budgets: determinism, watchdogs and graceful degradation.
- Linux, Docker, Git and CI as the natural working environment.
GPU & Inference Optimization (NVIDIA)
- Solid CUDA: execution model, streams, CUDA Graphs, memory management (pinned, unified, pre-allocation), writing and debugging custom kernels.
- GPU libraries: cuBLAS, NPP, CV-CUDA, Thrust or equivalents for accelerated image pre-processing and scoring.
- TensorRT in production: engine building, mixed FP16 / INT8 precision with custom quantization calibration, precision-degradation diagnosis and dynamic batching.
- Triton Inference Server in production.
- Profiling with Nsight Systems / Nsight Compute; p99 latency characterization per stage and finding the real bottleneck before optimizing.
Data & MLOps
- Model versioning and registry (MLflow or equivalent), reproducibility and dataset curation (CVAT).
- Traceability: able to demonstrate which model, data and version produced a given result.
- Models in production: monitoring, drift detection and a retraining / rollback policy.
Nice to Have
- Custom TensorRT plugins.
- Industrial cameras — GigE Vision (ideally Basler pylon); optics, lighting and photometric calibration (flat-field).
- Anomalib (advanced use or upstream contributions).
- Manufacturing / quality context (automotive or another regulated industry); ISA-95 and IEC 62443.
- Public cloud and cloud MLOps (ideally Azure: IoT Edge, ML).
- Publications, talks or open source in vision / anomaly detection.
¿Qué podemos ofrecerte:
- Formarás parte de un gran equipo de profesionales con inquietud y motivación por el desarrollo y la programación y participaras en nuevos y punteros proyectos para la compañia.
- Trabajarás a jornada completa con un horario flexible bajo un modelo de trabajo 100% remoto.
- 22 días de vacaciones + 2 de asuntos propios.
- Formación por parte de la empresa para que puedas seguir desarrollándote y promocionar dentro del plan de carrera que existe para ti.
- Podrás asistir a eventos y conferencias relevantes del sector.
- Contrato indefinido.
- Estabilidad y buen clima laboral.
- Salario Competitivo.
- Acceso a ventajas del grupo de empresa
- Retribución flexible y más beneficios
Skills Required
- Expert-level PyTorch, including CNN and transformer vision architectures, training, evaluation, and rigorous ONNX export
- Expertise in one-class and unsupervised anomaly detection, including PatchCore, EfficientAD, PaDiM, student-teacher methods, normalizing flows, threshold calibration, synthetic defects, and drift
- Experience with supervised defect segmentation and detection, including encoder-decoder and DETR-family models and imbalanced datasets
- Experience evaluating models using AUROC, AUPRO, escape rate, false-reject rate, and regression validation
- Expert-level modern C++17/20 and Python for real-time vision pipelines
- High-performance systems design involving capture, preprocessing, inference, post-processing, in-memory data handling, determinism, watchdogs, and graceful degradation
- Solid CUDA expertise, including execution models, streams, CUDA Graphs, memory management, custom kernels, and debugging
- Production experience with TensorRT, FP16/INT8 precision, quantization calibration, dynamic batching, and precision-degradation diagnosis
- Production experience with Triton Inference Server
- Experience profiling with Nsight Systems or Nsight Compute and characterizing p99 latency
- Experience with model versioning, registries, reproducibility, dataset curation, and traceability using MLflow or equivalent and CVAT
- Experience monitoring production models, detecting drift, and implementing retraining and rollback policies
- Working knowledge of Linux, Docker, Git, and CI
- Experience with custom TensorRT plugins
- Experience with industrial cameras, GigE Vision, Basler pylon, optics, lighting, or photometric calibration
- Advanced Anomalib experience or upstream contributions
- Manufacturing or quality experience, preferably automotive or regulated industry, with ISA-95 and IEC 62443
- Public cloud and cloud MLOps experience, preferably Azure IoT Edge and Azure Machine Learning
- Publications, talks, or open-source contributions in computer vision or anomaly detection
What We Do
Inetum is a European leader in digital services. Inetum’s team of 28,000 consultants and specialists strive every day to make a digital impact for businesses, public sector entities and society. Inetum’s solutions aim at contributing to its clients’ performance and innovation as well as the common good. Present in 19 countries with a dense network of sites, Inetum partners with major software publishers to meet the challenges of digital transformation with proximity and flexibility. Driven by its ambition for growth and scale, Inetum generated sales of 2.5 billion euros in 2023. Top Employer Europe 2024
.png){kind=logo}
