About this role
Your tasks
Your Tasks
- Drive the evolution of our recommendation landscape and shape how people discover employers, jobs and content on kununu.
- Own the full data science process: from framing the problem and exploring the data to modelling, evaluation and production rollout.
- Apply methods such as bandits and reinforcement learning to personalize content in a live, high-traffic environment.
- Design and build LLM-driven, agentic applications on top of a large body of unstructured data, and set up solid evaluation for them, including LLM-as-a-judge approaches.
- Work closely with product and engineering to turn business questions into data science problems and bring your models into the product.
- Define with product what success looks like: choose the right metrics and evaluation strategy, run experiments, and turn hypotheses into empirical results within weeks.
- ML-Lifecycle: Work with transformation tooling to build reproducible data foundations for your models, monitor quality and cost, and improve them over time.
- Mentor more junior colleagues in the team: review their work, share your knowledge and help them grow.
- Act as a sparring partner for the team lead on statistical methods, machine learning concepts and solution design.
Your profile
Your Skills
- Several years of hands-on data science experience, including at least one recommendation or personalization system you brought to production.
- A strong foundation in statistics and machine learning. Practical experience with multi-armed bandit experimentation or reinforcement learning is a big plus.
- You are comfortable across classical machine learning problems such as classification, ranking and scoring, not only recommendations.
- Real production experience with LLM-based applications: agentic tools, LLM judges and evaluation harnesses.
- Solid Python and SQL skills. You are comfortable in a modern data warehouse and with transformation tools such as dbt.
- You write clean, tested and reusable code, and you see version control and code reviews as a normal part of your work.
- You explain complex results in simple terms to people outside data science, and you enjoy mentoring and giving constructive feedback.
What we offer
Your Benefits
- Bring your dog to the office
- 6 week vacation
- Mobile devices also for private use
- Drinks, food & goodies
- Restaurant vouchers and employee discount
- Up to 12 weeks Workation
- Remote work option
- Trust-based working hours
- Communication on equal terms
Contact
Your Contact
You still have questions?
Feel free to get in touch with me.
Barbara
Bárbara Serrano
Recruiting Manager