2 months ago
Berlin, GermanySenior
Responsibilities
- Build and maintain end-to-end ML infrastructure, including automated training pipelines, model orchestration, and deployment tooling.
- Define and evolve model-serving standards and tooling for low-latency, highly available ML services.
- Develop self-service MLOps capabilities such as feature stores, model registries, and automated performance and data-drift monitoring.
- Enable Kubernetes autoscaling and GPU provisioning and operate a Kubernetes-based development cluster.
- Collaborate with Operations and ML teams to move models from experimentation into GPU-backed production.
- Design observability, define service-level objectives, participate in incident response, and automate repetitive operational work.
- Create standardized workflows and golden paths that help Data Scientists develop and deploy models efficiently.
- Modernize production infrastructure while managing reliability, risk, cost, availability, and continuity.
Requirements
- Experience building and operating machine learning platforms in production environments.
- Working knowledge of Docker, Kubernetes, Linux internals, and model serving at scale.
- Familiarity with ML lifecycle tooling, orchestration frameworks, feature stores, model registries, and drift or performance monitoring.
- Experience owning production systems, including SLOs, observability, incident response, and systematic diagnosis of large-scale failures.
- Production-quality programming experience in Python or a comparable language.
- Experience modernizing production infrastructure with attention to reliability, risk, cost, availability, and continuity.
- Ability to own technical outcomes, use data to support decisions, and communicate clearly with technical and non-technical audiences.
Benefits
- Hybrid work requiring commuting to the Berlin office three times per week.
- Work from almost anywhere for up to 20 days per year.
- Company-paid therapy sessions through SpringHealth and a company-paid HeadSpace subscription.
- Company-wide week off, no-meeting Fridays, paid parental leave, and paid volunteer time.
- Development Dollars, leadership development, and access to thousands of on-demand e-learnings.
- Travel discounts, Employee Resource Groups, six weeks of paid vacation, and a birthday day off.
- Free lunch two days per week, pension plan contributions, public transportation subsidies, and a bike leasing program.
- Monthly social events, Thursday happy hours, sports teams, and an office in Friedrichshain, Berlin.
Tech Stack
Categories
About KAYAK
KAYAK builds a travel metasearch platform that lets consumers compare flights, hotels, rental cars, and vacation packages across hundreds of sites, plus tools for trip planning. It earns through referrals and advertising and offers KAYAK for Business for corporate travel. Founded in 2004 and headquartered in Stamford, Connecticut, KAYAK is part of Booking Holdings and operates a portfolio of brands including momondo, Cheapflights, and HotelsCombined.