Senior Machine Learning Engineer

Senior Machine Learning Engineer – Data EnablementWho are we?Auction Technology Group (ATG) is transforming the multi-billion-dollar global auction industry. Our platforms connect thousands of auction houses with buyers in over 170 countries, powering more than $15 billion in annual sales. Through innovative online auction technologies, we help auctioneers expand their reach, boost efficiency, and maximize value—while giving bidders unrivaled access to rare and specialized items. As a publicly traded company, ATG has scaled from $18 million to $170 million in revenue, with sustained growth beyond the pandemic. We're modernizing one of the last industries to fully go digital—building a global, category-defining business in the process.Who are we looking for?We are making a significant investment in creating a user experience that meets the expectations of our customers. Not only do you put the customer at the heart of everything you do, but you are adept at enabling data-driven decisions to design and deliver strategic projects. You will be comfortable working cross-functionally with Product, Engineering, MLOps, and Analytics teams to develop our products and improve the end user experience. You should have a strong track record of successful prioritization, meeting critical deadlines and enthusiastically tackling challenges with an eye toward problem solving.What your contributions will be:Design and develop state-of-the-art recommendation algorithms leveraging collaborative filtering, content-based filtering, and hybrid approaches to surface relevant auction items to biddersBuild and optimize learning-to-rank models that re-rank search results and recommendations based on user preferences, behavioral signals, and contextual featuresDevelop personalization systems that adapt to individual user interests, browsing patterns, and bidding history across multiple auction categories and marketplacesBuild classification and embedding models to better represent our product taxonomy and enable semantic similarity matching across diverse auction itemsCollaborate closely with the engineering and MLOps teams to integrate machine learning algorithms into production systems and APIsPerform rigorous experimentation (A/B testing) to demonstrate the causal impact of recommendation strategies and conduct analyses to identify challenges and opportunities, deriving valuable insightsLeverage computer vision techniques to enhance visual similarity recommendations and improve content understandingStay updated with scientific advancements in recommender systems, personalization, and ranking, and contribute to technical publications when possibleWhat you need for Success:Educational Background:MSc or PhD in relevant fields such as Machine Learning, Data Science, Computer Science, Statistics, or related disciplines

Required Skills

Strong expertise in Python and familiarity with data science and machine learning libraries such as Pandas, NumPy, Scikit-learn, TensorFlow, and PyTorchSolid understanding of recommendation system architectures: collaborative filtering (matrix factorization, neural collaborative filtering), content-based filtering, and hybrid approachesExperience with learning-to-rank algorithms (e.g., pointwise, pairwise, and listwise approaches such as RankNet, LambdaMART, LambdaRank) and their application to re-ranking problemsProficient in deep learning techniques for recommendations, including neural networks, embeddings, two-tower models, and transformer-based architecturesUnderstanding of personalization techniques: user profiling, behavioral modeling, contextual bandits, and online learningExperience with evaluation metrics for recommender systems (e.g., Precision@K, Recall@K, NDCG, MRR, diversity metrics, coverage)Familiarity with handling sparse data, cold-start problems, and implicit feedback signalsKnowledge of feature engineering for recommendation systems, including user features, item features, and interaction featuresUnderstanding of A/B testing frameworks and experimental design for measuring recommendation qualityNice-to-Have:Experience with large-scale embedding systems and vector databases (e.g., Elastic, Milvus, Pinecone)Familiarity with computer vision models for visual similarity and image-based recommendationsKnowledge of multi-armed bandit algorithms and exploration-exploitation strategiesExperience with session-based or sequence-aware recommendation models (e.g., RNNs, transformers for sequential recommendations)Understanding of fairness, diversity, and serendipity in recommendation systemsExperience with marketplace or e-commerce recommendation systemsSoft Skills:Ability to conduct practical research with a scientific mindset and a focus on delivering actionable resultsStrong communication and interpersonal skills, with a proven ability to work collaboratively in a team-oriented environmentExcellent problem-solving skills, capable of abstracting complex problems into their essential components and developing effective solutionsAbility to balance technical excellence with business impact and user experience considerations