• Design, build, and operate the online feature store — the real-time serving layer that delivers ML features to production models with low latency and strong consistency guarantees • Build and maintain data pipelines (batch and streaming) that compute, validate, and publish features from source systems into the online and offline stores • Work on the data models and Postgres schemas that back real-time feature serving, optimizing for query performance, freshness, and scale • Act as a technical leader for feature store infrastructure; help drive enhancements to quality, scalability, reliability, and observability for the systems ML models depend on • Partner closely with our Data Science, Analytics Engineering, Product Management, and Application Development teams to translate business requirements into production-grade engineering solutions • Contribute to mentoring for junior engineers • Contributing to best practices and raising the bar through thoughtful code reviews and contributions to technical design discussions

• Construção de pipelines de Machine Learning de ponta a ponta: incluindo preparação de dados, otimização de código, automatização de rotinas de treino e predição de modelos; • Monitoramento de métricas de qualidade, eficiência, custo e uso de produtos de dados; • Liderança técnica no processo de MLOps do time, definindo boas práticas de desenvolvimento e sustentação de soluções; • Desenvolvimento de arquiteturas em nuvem para soluções de dados e IA (GCP, AWS, Azure); • Construção de pipelines de dados com modelos de Gen AI embarcados (ex: preparação de dados não estruturados para estruturados); • Desenvolvimento de pilotos junto à área para prototipação de conceitos e prova de valor.

• Own the full lifecycle of predictive models in production — architecture, training pipelines, inference infrastructure, deployment, and ongoing model health • Build and operate the systems that route model outputs into live product surfaces: search ranking, recommendations, feed ordering, and related user-facing experiences • Establish and maintain model monitoring, alerting, drift detection, and retraining cadences — the feedback loops that keep deployed models accurate over time • Partner closely with Data Science, Data Engineering, Product Management, and backend engineering to move work from validated approach to production system • Own the decision-making process on whether to leverage ML infrastructure & expertise from our parent company, GOAT Group, and when to advocate for building in-house solutions. • Contribute to ML infrastructure decisions — serving architecture, feature computation, pipeline orchestration — with an eye toward what scales as the team and model count grows • Set technical standards and raise the bar for how ML systems are built, evaluated, and operated across the pod

• Design, develop, optimize and deploy ML models, including large language models, for various natural language processing tasks. • Implement and maintain training pipelines, leveraging distributed training and optimizing for performance and efficiency. • Collaborate with cross-functional teams to gather requirements, define model architectures and iterate on model development. • Conduct model evaluations, performance analysis, and optimization to improve model accuracy and reduce biases. • Stay up-to-date with the latest research and advancements in the field of natural language processing, multimodal signals, and large language models. • Contribute to the development of best practices, guidelines, and ethical AI principles for responsible ML development and deployment.