Machine Learning Engineer — One Matching
Team: One Matching · Location: Prague / Czech Republic (hybrid) · Level: Mid-level, with room to grow
About the Team
One Matching connects offers from thousands of shops to the right products in our catalogue, across Heureka's marketplaces. We build and run the systems that do it, at the scale of millions of offers.
What we mean by "ML Engineer"
For us, an ML engineer works across four areas: software engineering, model training, evaluation, and data analysis. You don't need to be the best at all four, but you should be able to move between them.
What You Will Do
- Own problems end to end, from understanding the business need to measuring the result in production.
- Build and improve ranking, matching, and data-quality systems, using both classic ML and LLMs.
- Work with content and product teams to turn vague business problems into concrete technical plans, and explain the trade-offs.
- Help decisions along by proposing a direction and working with the team to reach a solution.
- Think about the whole system, from search and features through models, thresholds and deployment, and the services that depend on them.
What We Are Looking For
- You think business-first and weigh effort against value.
- Applied ML experience: training and evaluating classic models (XGBoost, scikit-learn), calibration, error analysis, and knowing when not to use ML.
- Hands-on MLOps, from deploying models to monitoring and maintaining them in production.
- Solid software engineering, including reliable production Python you own end to end. Cloud experience, ideally GCP.
- You can scope work, sequence it, and estimate it realistically.
- Enough ownership to decide with incomplete information and correct course as you learn.
- Clear communication with engineers and non-technical stakeholders.
Nice to Have
- LLMs in production (annotation, evaluation, extraction)
- Elasticsearch or similar
- Vertex AI, our ML pipeline platform
- E-commerce experience
Our Stack
Python, GCP, Elasticsearch, XGBoost, classic ML and LLMs. LLMs are in production today, and we are moving toward agent-based development.
How We Work
We are a small team, so you get a lot of say in what you work on and how you do it. You will not be doing it alone. We plan together, review each other's designs, and help each other get things into production. We keep solutions as simple as the problem allows.
What We Offer
- Impact: Your code shapes the shopping experience for millions of people.
- Hackathons & conferences: We support your participation in events like DevFest or devopsdays.
- Flexibility: Home office up to 3 days a week and flexible hours. We meet in the office at least once every two weeks for team collaboration and scrum ceremonies.
- Career journey: A clear roadmap to help you grow.
- The basics: 25 days of vacation, extra time off, a Cafeteria benefit system, and a Multisport card.
