Amazon
Senior
Own testing/monitoring infrastructure for EC2's largest AI/ML network stacks
Interview for a Senior Software Engineer role on Annapurna Labs' ML Network Stack team, which supports frameworks like NCCL, NVSHMEM, and NIXL for EC2's distributed AI/ML systems. The role focuses on building the infrastructure that monitors, benchmarks, and reports on performance/functional regressions across massive-scale testing workloads.
Step into this interview
Free · a live voice mock calibrated to this exact role
What this interview tests
- Networking and HPC/RDMA interconnects for distributed AI/ML systems
- Building automated testing/benchmarking infrastructure at scale (Python, CI/CD)
- Performance data pipelines and dashboarding (Grafana, Athena)
- Automatic regression/anomaly detection across instance types and software stacks
- SW/HW co-design and technical leadership/mentoring
Common question themes
Design a system to automatically spin up large ML/HPC clusters, run benchmarks, and detect performance regressions before they reach customers
Walk through your experience with NCCL, RDMA, or high-speed networking internals in a distributed ML context
How would you build a dashboard pipeline from raw performance telemetry (e.g., via Grafana/Athena) to actionable developer alerts
Describe a SW/HW co-design tradeoff you made and how it affected system performance or reliability
Tell me about leading or mentoring a team through building test infrastructure that had to scale across many instance types and OS versions
Related interviews
Amazon
Mid
Software Development Engineer II, Amazon Smart Vehicles
Amazon
Mid
Software Development Engineer, Ads AI Core Infra
Amazon
Mid
Software Engineer II, Leo Regulus
Figma
Mid
Software Engineer, Traffic
Airbnb
Mid
Software Engineer, Cloud Networking
Cloudflare
Mid