Rajeev Kumar Ainampudi

About Rajeev Kumar Ainampudi

Client-focused Azure Certified Engineer with 20+ years of experience in software engineering, specializing in application support, cloud infrastructure, CSS, and DevOps. Skilled in Azure AD, Entra ID, Microsoft Sentinel, Windows, SQL, and automation using PowerShell, Git, Jenkins, Docker, and Ansible. Adept at system monitoring, diagnostics, and log analysis across cloud and hybrid environments. Known for clear communication, strong documentation, and analytical problem-solving. Committed to customer success through technical ownership, continuous learning, and cross-functional collaboration to deliver secure, scalable, and resilient solutions.

Experience

1. Gpu Node Recovery Engineer Azure Cloud Support Engineer

   LTIMindtree

   Jun 2024 — Present · Redmond, WA, US

   LTIMindtree Ltd | Microsoft – Redmond, WA Jun 2024 – PresentProject: GPU FW COE Team Support• Diagnosing and troubleshooting node issues by identifying root causes and restoring systems affected by diverse error conditions.• Create and secure secrets and encrypt passwords using Azure Key Vault.• Leverage Azure Storage Explorer (ASE) to efficiently access and gather node details.• Employ Azure Monitor to detect and resolve performance issues and utilize Azure Advisor to generate cost optimization recommendations based on underutilized Azure VMs.• Manage Azure Privileged Identity Management (PIM) for Just-In-Time (JIT) access.• Analyze diagnostic logs, including GPU logs, CSI diag logs, PCIe tree logs, and Nvidia tool outputs.• Utilize Kusto Explorer to craft advanced queries for retrieving node data based on node state across each FC of the GPU.• Collaborate with DCM Explorer to recover nodes and transition them back into the production pool.• Execute updates to GPU bundle fwpkg, baseboard retimer, and MB retimers using Out-of-Band (OOB) methods via Redfish curl commands.• Update firmware on DC-SCM components, including BIOS PFM, BMC PFM, CFM, PCD, and ME of BIOS using OOB techniques.• Investigate and document repeat offender nodes and unsupported tickets and provide repair actions for DcOps teams.• Resolve BMC link-down faults on servers in collaboration with DcOps during work-window sessions.• Utilize Azure Sentinel (SIEM) for threat detection.• Perform in-band updates and validation of new CFM on GEN 9.3 and 9.4 H100 systems, including rollback and upgrade of GPU bundle fwpkg versions (1.3.4 and 1.5.0), and validate component attestation logs.• Execute OaaS scripts to update GPU firmware and transition nodes back to production.• Collect and analyze GPU diagnostic dump logs (Retimer, FPGA, Erot, HMC, Self-test), along with CPER, journal logs, and baseboard log entries via Redfish commands.

Education

• University of Madras

  Master of Computer Applications - MCA