Data Scientist

A high-impact AutoTech and Connected Mobility AI company developing an intelligent telematics and predictive analytics platform for the global automotive industry.By ingesting and analyzing billions of real-time vehicular data signals, the platform enables real-time anomaly detection, predictive maintenance, and driver behavioral profiling for global automotive OEMs, Tier-1 suppliers, and fleet operators.Operating its core R&D center in Israel, the organization combines massive Big Data streaming infrastructure with production-grade Machine Learning and Agentic AI systems.Located in the Sharon region (Herzliya, walking distance from the train station), operating on a hybrid work model with 1 day WFH per week.Role Description-Serving as a Senior Machine Learning / AI Production Engineer, taking full end-to-end ownership of the ML lifecycle—from exploratory data analysis and custom architecture modeling to high-throughput production deployment and ongoing operations.Processing and modeling billions of noisy, high-velocity, real-time vehicular telemetry data points to build robust predictive maintenance and behavioral models.Designing, training, and optimizing custom deep learning architectures in PyTorch, transitioning complex prototypes into scalable, testable production services.Taking complete operational ownership of models in Production, including continuous monitoring, automated retraining pipelines, drift detection, and live debugging under robust CI/CD frameworks.Developing and deploying production-grade LLM and Agentic AI applications, implementing evaluation frameworks (Evals), guardrails, latency optimization, and cost-governance mechanisms.Collaborating cross-functionally with Big Data Engineers, Backend Architects, and Automotive Product domain experts.Requirements-5+ years of hands-on industry experience as a Machine Learning Engineer or Data Scientist – MandatoryProven track record of taking ML / Deep Learning models fully into Production with ongoing operational ownership (monitoring, drift mitigation, retraining, debugging; prototype-only experience is not sufficient) – MandatoryDeep proficiency in Python, with proven experience authoring tested, enterprise-grade production modules and/or owning microservices under CI/CD – MandatoryExtensive hands-on experience building and training custom architectures using PyTorch and deploying them to production – MandatorySolid hands-on experience working within Big Data Platforms & distributed data infrastructures – MandatoryDirect ownership of Production Engineering workflows / MLOps pipelines – MandatoryHands-on experience deploying LLMs or Agentic AI applications in Production (including evaluations, guardrails, and latency/cost tuning) – Significant AdvantagePractical experience working with Noisy, Unstructured, or Weakly-Labelled Data / Time-Series Telemetry – Significant Advantage