Teams at SumerSports

Build and operate scalable data pipelines supporting batch and streaming workloads, AI and ML systems, analytics, and sports intelligence products. Responsibilities include ETL/ELT development, data modeling, orchestration, retrieval pipelines, data quality, lineage, monitoring, infrastructure optimization, and collaboration with ML, product, and analytics teams.

Build and improve computer vision models for sports video, including detection, tracking, pose estimation, event recognition, and identity association. Lead experimentation, evaluation, error analysis, and data-efficiency improvements. Collaborate on dataset design and help productionize models through packaging, inference optimization, automated evaluation, and quality monitoring. Strong Python and PyTorch skills are required, with video computer vision experience preferred.

Designs and operates scalable ML infrastructure, including training, serving, inference, orchestration, and deployment pipelines. Optimizes GPU and CPU workloads, manages cloud clusters, implements observability and SLOs, and secures automated CI/CD and infrastructure-as-code workflows. Partners with research and AI engineering teams to productionize models, while building reusable tooling, templates, and runbooks. The role also supports low-latency model serving, data quality, model promotion, and cost optimization.