Job
- Level
- Junior
- Job Field
- Data
- Employment Type
- Full Time
- Contract Type
- Permanent employment
- Location
- Karlsruhe
- Working Model
- Hybrid, Onsite
Job Summary
In this role, you will enhance ML-based forecasting models, perform feature engineering, collaborate closely using Git, and deploy your analyses into production on AWS services.
Job Technologies
Your role in the team
- You continuously develop our ML-based forecast models through feature engineering, transfer learning, time series analysis, and the targeted use of new APIs.
- Clean, maintainable production code is your top priority. You focus on high code quality and enjoy conducting code reviews together with your colleagues. Unit tests and test coverage are not foreign concepts to you.
- On the technical side, you work with Git and have the necessary structure to collaborate with the team on code. Basic knowledge of databases – whether SQL, NoSQL, or both – enables you to query and prepare data independently.
- You actively contribute to the joint development of new products and solutions. In collaboration with our CloudOps team, these are deployed to production on AWS services.
This text has been machine translated. Show original
Our expectations of you
Education
- You have a completed degree in Data Science, Computer Science, Mathematics, Statistics, Physics, or a comparable quantitative field, and you bring initial experience in the Data Science environment — for example, with Machine Learning in Forecasting or Recommendation Systems.
Qualifications
- You work passionately with Python. Object-oriented programming is second nature to you, and you are familiar with libraries such as pandas, NumPy, and scikit-learn. You know how to process data efficiently, build models, and do all of this in clean, maintainable code.
- Feature engineering sounds exciting to you: You can derive meaningful features from raw data, whether from timestamps, categorical variables, or external signals.
- You have a basic understanding of time series analysis: you are familiar with the concepts behind seasonality, trends, and autocorrelation, and have ideally trained initial models (e.g., ARIMA or ML-based approaches).
- You have a solid statistical foundation — hypothesis testing, distributions, and confidence intervals are familiar to you, and you can interpret results confidently and work in a methodologically sound manner. In exploratory data analysis, you can quickly navigate unfamiliar datasets, identify patterns, anomalies, and data quality issues.
- You speak very good German, have solid English skills, and are willing to come to Karlsruhe approximately once a month (more often during the onboarding period).
Experience
- Ideally, you have experience with AutoML frameworks (e.g., AutoGluon) or time series libraries such as GluonTS, Darts, or NeuralForecast. Knowledge of Transfer Learning or Domain Adaptation, a basic understanding of Docker and cloud services, as well as experience with data pipelines or workflow orchestration, are a plus.
This text has been machine translated. Show original
What we offer
- Look forward to trust-based leave, flexible working hours in a modern office building, and the opportunity to work remotely.
- We pull together and give each other a mutual trust boost. We don't look for faults but for solutions and celebrate our successes together.
- Our highly motivated team shares a passion for our collective work and our products. With great team spirit, creativity, and enthusiasm, we pursue our goals together. Passion is our secret code.
- As a motivated team member, you can significantly shape our culture and our products at Nesto, and together as a team, achieve great things for our customers.
This text has been machine translated. Show original
Topics that you deal with on the job
Job Locations
This is your employer
Nesto Software GmbH
Nesto ermöglicht eine intelligente und bedarfsorientierte Personaleinsatzplanung für die Systemgastronomie. Mithilfe selbstlernender Algorithmen prognostiziert Nesto automatisch den Personalbedarf für zukünftige Tage und Stunden und macht so Über- sowie Unterbesetzungen bereits bei der Erstellung der Dienstpläne sichtbar. Die integrierte Analyse der Umsätze in Echtzeit erkennt und warnt rechtzeitig vor bevorstehenden Personalengpässen oder Überbesetzungen – bevor diese überhaupt eintreten. So ermöglicht Nesto seinen Kunden, die Personalauslastung zu optimieren sowie die Planungssicherheit zu erhöhen.
Description
- Company Size
- 1-49 Employees
- Company Type
- Established Company
- Working Model
- Hybrid, Onsite
- Industry
- Internet, IT, Telecommunication
Junior Data Scientist - Forecasting
- Location
- Karlsruhe
- Working Model
- Hybrid, Onsite
- Diversity
- Open for all genders
