Teams at Andromeda (andromeda.ai)

Own technical qualification of compute providers: define acceptance standards and benchmarks, validate clusters (fabric, GPUs, storage, orchestration), guide provider remediation, and maintain technical relationships to ensure reliable production performance.

Embed with customer teams running large-scale GPU training and inference to onboard, tune, debug, and improve reliability. Diagnose fabric, driver, scheduler, and application failures; profile performance; build automation, monitoring, and preflight checks; lead incident response; and convert field learnings into product improvements and reusable reference configurations.

Own technical engagements end-to-end for large customers: lead technical discovery and POCs, design GPU/HPC cluster and workload architectures, provision and validate production training runs, optimize performance and cost, and build SA assets (runbooks, benchmarks, reference architectures). Act as named technical owner to grow accounts and feed product/engineering with customer feedback.