Senior AI/ML Engineer

Bloomreach is building the world’s premier  agentic platform for personalization .We’re revolutionizing how businesses connect with their customers, building and deploying AI agents to personalize the  entire  customer journey. We're taking  autonomous search  mainstream, making product discovery more intuitive and conversational for customers, and more profitable for businesses. We’re making  conversational shopping  a reality, connecting every shopper with tailored guidance and product expertise — available on demand, at every touchpoint in their journey. We're designing the future of  autonomous marketing , taking the work out of workflows, and reclaiming the creative, strategic, and customer-first work marketers were always meant to do. And we're building all of that on the intelligence of a single AI engine —  Loomi — so that personalization isn't only autonomous…it's also consistent.From retail to financial services, hospitality to gaming, businesses use Bloomreach to drive higher growth and lasting loyalty. We power personalization for more than 1,400 global brands, including American Eagle, Sonepar, and Pandora. You'd be joining the  Artificial Intelligence team . We own the algorithmic core of the platform: Predictions, Contextual Personalization, contextual bandits, autosegmentation, and the agentic workflows behind Loomi. We work with behavioural data at terabyte scale, across 1,400+ customers, in production, every day. You'll work on cutting-edge technologies, impacting millions of users, and contributing to a product that truly makes a difference. Working in one of our  Central European offices (Bratislava, Brno, Prague) or from home (Czechia, Slovakia) on a full-time basis , you´ll become a core part of the Engineering Team . The mission You turn a model that works in an experiment into a service that works for 1,400 customers. You own ML-powered features end to end — the API that configures them, the pipeline that trains them, the endpoint that serves them, and the monitoring that tells you when they've drifted. That includes L3 escalations on what you ship. We think engineers who never see a production incident build worse systems. What you'll actually do Own features end to end — back-office APIs, training and inference pipelines, high-performance serving endpoints, monitoring. Take models from the modelling side and make them production-grade. Usually that means rewriting the training path, thinking hard about feature freshness, and finding out what a model does with inputs nobody anticipated. Design for the scale we actually run at — terabytes of behavioural data, multi-tenant, with latency budgets that make the naive approach unworkable. Build the quality gates — unit and integration tests, shadow deployments, A/B testing, drift and performance alerting. Noticing a model has quietly degraded is your problem. Support what you ship, including L3 investigations with our client-facing colleagues. Write things down — design docs, runbooks, decision records. The team is growing and undocumented systems don't scale with it. This role bends in three directions We'd rather shape it around you than the other way round. ML product engineering — you own an ML-powered feature end to end and the ML part is what makes it interesting. Platform & MLOps — pipelines, serving, deployment, observability, inference cost. You want the system, not the model. Modelling & data science — features, experiments, and whether the number means anything. Most people lean one way and dip into the others. This opening is centred on the first two. If modelling is where you'd want to spend most of your time, our  Senior Data Scientist opening (in Czechia or Slovakia) is the better fit, and applying to both is fine. What you'll need 5+ years of Python engineering. You build production services, not scripts, and you have opinions about testing them. Real experience running ML in production — serving, retraining, monitoring, and the specific ways ML systems fail that ordinary services don't. Solid software architecture fundamentals: scalable APIs, microservices, data flow design. Cloud platform depth — we run on GCP and Databricks ; AWS or Azure background transfers fine. Plus Git and CI/CD as a matter of course. Working English, written and spoken. Plus real depth in at least one of: Kubernetes, CI/CD and infrastructure-as-code — the platform side. Distributed data processing: Spark on Databricks, Dataproc, or equivalent. Feature stores, model registries, or experiment tracking tooling — MLflow, Unity Catalog, Vertex AI Model Registry. Evaluation and experiment design — you know why a metric moved, and when it didn't move for the reason everyone assumes. Also good: you use agentic coding tools daily and have a view on where they help and where they quietly don't. We build agents for a living, and people who use them tend