Summary
Alexey Kipnis is an AI Infrastructure and MLOps engineer with 8+ years of experience building reliable, automated platforms for data-driven products. He has migrated complex services to Kubernetes at scale and designed reproducible ML pipelines with DVC and MLflow, replacing brittle local workflows with tracked, versioned training and deployment flows. At CodeValue he deployed Triton + KServe for GPU inference and implemented a fractional/dedicated GPU policy that reduced latency variance while improving utilization, complementing this with Prometheus/Grafana and DCGM-based observability and auto-rollback. His background spans system administration, security-aware DevOps, and high-load web applications, giving him practical strengths in CI/CD, Helm/ArgoCD, and pipeline templating. Based in Israel with an MS in Computer Science, he blends infrastructure pragmatism with an emphasis on measurable, production-grade ML delivery.
8 years of coding experience
14 years of employment as a software developer
Master's degree, Computer Science, magister, Master's degree, Computer Science, magister at South Ural State University (SUSU)
English, Hebrew, Russian