Senior Data Scientist

AB InBev is the leading global brewer and one of the world’s top 5 consumer product companies. With over 500 beer brands, we’re number one or two in many of the world’s top beer markets, including North America, Latin America, Europe, Asia, and Africa. About AB InBev Growth Group Created in 2022, the Growth Group unifies our business-to-business (B2B), direct-to-consumer (DTC), Sales & Distribution, and Marketing teams. By bringing together global tech and commercial functions, the Growth Group allows us to fully leverage data and drive digital transformation and organic growth for AB InBev around the world. In addition to supporting well known global beer brands like Corona, Budweiser and Michelob Ultra, the Growth Group is home to a robust suite of digital products including our B2B digital commerce platform BEES, on-demand delivery services Ze Delivery and TaDa Delivery, and tabletop beer keg PerfectDraft. We are an exceptional team, focused on understanding and supporting consumer and customer needs, harnessing new technology, and scaling growth opportunities. About the Role The Senior Data Scientist will be responsible for developing solutions at the intersection of Computer Vision, Computer Graphics, and Multimodal Models. This role involves tackling strategic challenges, combining traditional computer vision techniques with cutting-edge scene reconstruction and generative AI, focusing on detection, classification, visual similarity, synthetic data generation, and multimodal interpretation. You will work in a highly collaborative, global environment, transforming visual data into robust and scalable models, ensuring automation, efficiency, and high performance across implemented pipelines. We are looking for someone with an analytical mindset, curiosity, a strong technical foundation in the mentioned areas, excellent communication skills, a collaborative spirit, and a passion for innovation and continuous learning. What You Will Do Fine-tune open-source object detection models (e.g., YOLO, RT-DETR, and similar architectures), applying them to real-world problems with an emphasis on performance, accuracy, and scalability. Develop and adapt image classification models, with experience in architectures such as EfficientNet, MobileNet, and ResNet, focusing on production readiness. Work with Siamese networks for verification and visual similarity tasks, applying fine-tuning and optimization strategies for image comparison and visual integrity validation. Apply and customize multimodal generative and instruction-following models such as Qwen-VL, PaliGemma, and BLIP, using fine-tuning techniques like LoRA and QLoRA, focusing on image-text alignment. Have knowledge of scene reconstruction techniques such as NeRF or Gaussian Splatting, and hands-on experience with frameworks like Nerfstudio or similar. Work with large volumes of visual data, coordinating curation, annotation, and qualitative evaluation strategies to ensure diversity and quality in image datasets. Generate and use synthetic and 3D data as complementary sources for training and evaluating models, supporting their creation and curation to enhance model generalization and robustness. Automate, version, and monitor training, inference, and deployment pipelines using Python, PySpark, MLflow, GitHub Actions, Docker, and CI/CD, running on cloud infrastructure (especially Azure and Databricks). Collaborate with data engineers, data scientists, PMs, and business stakeholders, ensuring technical alignment and strategic impact of developed solutions. Work in an agile, global environment, participating in Scrum/Kanban ceremonies, technical reviews, and continuous delivery and learning cycles. What You Need to Have Bachelor’s degree in engineering, Computer Science, Mathematics, Statistics, Operations Research, or related fields. A master’s or PhD in a relevant area is a plus. General and foundational knowledge in data science and statistics, including data analysis, modeling, and validation techniques. Strong background in Deep Learning, including mastery of Transformer architectures and advanced algorithms applied to real-world problems. Hands‑on experience building models for object detection, image classification, visual information extraction, and multimodal interpretation. Proficiency in TensorFlow or PyTorch, with experience in model development and deployment. Experience with cloud computing, particularly Azure, Databricks, and Spark, for large‑scale data processing. Expertise in Python and PySpark, plus familiarity with data manipulation and modeling libraries and frameworks. Knowledge of version control and collaboration tools such as GitHub, and experience with CI/CD pipelines for data and model projects. Understanding of automation workflows, model monitoring, and best practices for maintaining performance and robustness. Advanced English to operate in a global environment. Strong interpersonal skills, communication, teamwork, organization, autonomy, and focus on delivering high‑quality results. What We Offer: Performance based bonus* Attendance Bonus* Casual office and dress code Days off* Health, dental, and life insurance Discounts on Ambev products* School materials assurance Language and training platforms Equal Opportunity & Affirmative Action: AB InBev Growth Group is proud to be an Equal Opportunity and Affi­x­ative Action employer. We do not discriminate based upon race, color, national origin, gender, gender identity, sexual orientation, protected veteran status, disability, age, or other applicable legally protected characteristics. #J-18808-Ljbffr