Lead Data Scientist

Stellenbeschreibung:

Overview

Digis is a European company of 200 employees. The company was founded 9 years ago and has grown from a startup into a reliable partner for our clients. You collaborate directly with the client's team, make business trips....

About Digis

About Digis
Digis is a European IT company with 200+ specialists delivering complex SaaS products, enterprise solutions, and AI-powered platforms worldwide. We ensure transparency, stability, and professional growth opportunities for all our team members.

About the Project

About the Project
Anext-generation simulation and analytics platform for industrial utilities, designed to transform real-time sensor data into operational insights. The platform combines live field input, statistical inference, and event data to deliver explainable recommendations that improve performance, reduce maintenance costs, and optimize resource usage. This is a high-impact ML project—early-stage, research-driven, and focused on solving real engineering challenges, not just building LLM-based prototypes. You’ll join as a founding technical contributor and have long-term ownership over the analytics layer and modeling strategy.

Stack

Stack
Python, ML frameworks (PyTorch, TensorFlow, scikit-learn)
Data analytics & simulation tools, sensor data pipelines

Requirements

5+ years of commercial experience with Python and modern ML frameworks (PyTorch/TensorFlow/scikit-learn).
2+ years of experience building and operating production ML pipelines.
2+ years of experience with physics-informed or hybrid models and/or asset reliability analytics.
Hands-on experience with time-series sensor data (anomaly detection, failure prediction).
Experience in water networks or similar process industries (highly desirable).

Responsibilities

Lead the design and development of physics-informed and stochastic models for asset reliability and risk assessment.
Build and validate machine learning models for anomaly detection, failure prediction, and system optimization.
Analyze and interpret sensor and operational data to identify patterns and derive actionable insights.
Implement data preprocessing, feature engineering, model training, validation, and explainability workflows using Python and modern ML frameworks.
Collaborate with engineers, domain scientists, and product stakeholders to integrate analytics into digital twin and simulation platforms.
Develop frameworks for uncertainty quantification, causal inference, and probabilistic reasoning to improve decision-making under uncertainty.
Provide technical leadership, mentor junior data scientists, and establish best practices for scalable, production-grade ML pipelines.