
Senior AI Engineer - Harness Engineering (Kimchi)
Build and evolve the agent harness used to run LLM-based agents on CAST AI's inference infrastructure. Responsibilities include designing and shipping harness features (hooks, permission gates, context compaction), owning long-horizon multi-session execution, architecting planner/executor/evaluator pipelines, managing agent memory and state persistence, and owning the harness surface (CLI/TUI/MCP/telemetry). Tech stack includes TypeScript, Go, Kubernetes, cloud providers (AWS/GCP/Azure), Prometheus/Grafana/Loki, GitLab CI, and ArgoCD. Role is remote-first and open to candidates residing in specific European countries listed in the posting.










