About the Role
Wolt’s Personalization team is responsible for creating a tailored experience for Wolt’s customers across their shopping journey, selecting the best restaurants, dishes or items to match their culinary and shopping preferences across multiple premises such as Discovery, In Venue or Checkout. The Personalization team owns the ML stack, models and integrations that generate real-time recommendations for millions of customers across all Wolt markets. For example, the team is responsible for the models that rank restaurants in Wolt’s Discovery and Restaurants tabs, venues in Wolt’s Discovery and Stores tabs or items in Wolt’s in-venue and cart premises. It also personalises other components in Wolt’s shopping experience such as brands, banners or food categories that are relevant entry points for our customers to explore Wolt’s assortment.
Responsibilities
- Build the ML infrastructure to develop, train and deploy Wolt’s ranking models that select the content to display to our customers;
- Work end-to-end, from use case design to implementation, delivery and monitoring of your solutions;
- Maintain our production ML stack and raise the team’s ML engineering excellence bar;
- Liaise with Wolt’s ML Platform team to adopt different ML technologies and to create technical requirements for their solutions;
- Contribute to Wolt ML Engineering and Applied Science communities;
- Be part of a cross-disciplinary team with Applied Scientists, Software Engineers and Analysts to provide solutions to customer problems with a direct impact on the company’s business KPIs;
- Work at Wolt’s scale: Wolt operates in 30 different markets with millions of customers.
Qualifications
- You are experienced in end-to-end machine learning deployments and maintenance of ML systems and have at least 2+ years of experience in ML/MLOps;
- You have deployed and ran ML models in production at scale, maybe with hundreds of RPS and low latency;
- You bring solid experience in scaling solutions, monitoring ML stacks and troubleshooting ML deployments to the table. You can help with the technical issues the teams encounter;
- You are experienced in implementing real-time inference ML models in production;
- Good understanding of ML and MLOps principles as well as Software engineering experience in Python should complete your profile;
- Experienced in Docker, Kubernetes, workflow orchestration tools (e.g. Flyte), model and experiment registries (e.g. MLflow) and model serving systems (e.g. Seldon);
- You have solid communication and collaboration skills and are experienced in coordinating initiatives with your team and main stakeholders.