● Oversee the deployment and configuration of AI and HPC systems, including setting up operating systems, drivers, and optimizing server with GPU for maximum performance● Designed and architect hardware and software infrastructure for AI and HPC applications, including selecting and configuring servers, storage, and network components● Diagnosed and resolved hardware and software issues that may arise, ensuring minimal disruption to ongoing AI and HPC operations and conducted performance benchmarking and tuning to maximize the efficiency of AI models and HPC workloads● Linux Plumbing & Kernel Eng: Maintain Linux kernel and core userspace subsystems including submitting patches upstream against latest stable releases, with a focus on networking, and optimizing performance on Linux hosts with dozens to hundreds of CPU cores● Work closely with our internal customers, software developers, and other stakeholders to align kernel development with overall project goals, including customer requirements for GPU-based solutions and other specialized needs.● Leveraged Ansible and Bash scripting to manage over 3000 remote servers from a centralized jump box, maintaining system architecture using Grafana for performance monitoring and continuously monitor system performance and health, using tools and techniques to identify and address issues proactively.● Supervised and managed processes and applications, including Group Policy, ADDS, and client systems across 2900 devices. Performed migration/upgrade from RHEL 7 to RHEL 8 and installed Certbot on Linux open source and managed it● Proficient in configuring and managing storage solutions using logical Volume Manager (LVM) to optimize disk space allocation, create logical volumes, and implement snapshots for data backup and recovery purposes● Configured and optimized DNS servers for efficient resolution and load distribution and implemented DNS security measures such as DNSSEC and DNS firewall to enhance system protection.

 Identified and resolved performance bottlenecks in the Linux and AWS networking stacks, optimizing network traffic for containers in a Kubernetes infrastructure to enable efficient scaling and lower networking costs. Optimizing entire server fleet to get every last usable CPU cycle executing on our latency-sensitive and throughput-sensitive workloads. Led the technical design, development, and delivery of new features and in the resolution of critical software related issues and build the data and coordination systems that enable ultra-long context inference and training on GPU clusters Deployed, configured and maintained GPU-based compute infrastructure and software infrastructure for AI and HPC applications, including servers, storage, networking and associated software stack as well as providing detailed specifications and schematics for hardware selection and capacity planning. Integration: Ensured seamless integration of system components to create an efficient environment for executing complex computations and conducted performance benchmarking and tuning to maximize the efficiency of AI models and HPC workloads Implemented load balancing and resource management strategies to distribute computational tasks effectively and continuously monitor system performance and health, using tools and techniques to identify and address issues proactively. Leveraged Ansible playbook and Bash scripting to manage 4000+ remote servers from a centralized jump box, maintaining system architecture using Grafana for performance monitoring. Worked closely with AI researchers, data scientists, and HPC developers to understand their requirements and provide tailored solutions that support their work as well as maintaining thorough documentation of system configurations, procedures, and troubleshooting guides to facilitate knowledge sharing and ensure consistency Partnered with cross-functional teams to architect and manage infrastructure-as-code and cloud environments

• Installed, configured, and managed proprietary applications on Unix/Linux servers• Created and extend physical volumes, volume groups and resized existing logical volumes for additional space requirements• Use Hyper-V to manage memory, CPU, disk size and networking on windows servers• Troubleshooting network on Windows servers and adding peripheral devices on servers as needed and run security checks on windows systems, user account management and group collaboration• Administered Red Hat Systems and managed backups and monitored CPU and Disk Usage• Monitored SU logs for unauthorized root usage and access, monitored Unix/Linux servers load average, iostat, vmstats and general server health• Used remote tools such as remote desktop, Scap-workbench Utility to analyze and resolve issues in multiple servers and perform system monitoring; CPU, memory, I/O, hardware, jobs scheduling and processes management• Created and managed user/groups, set password, permissions, account expiration, reset user passwords, administered users account security through monitoring login logs• Restricted access to files and directories using Access Control Lists file permissions and utilized the Vi editor in accessing file contents and securely modifying them to meet specific objectives• Hardening Red Hat Linux 7 server and managed patching on Red Hat 7 server using yum• Monitored client disk quotas and general disk space usage – system performance monitoring and tuning and use RAID as needed• Developed core code modules, unit test tools and release notes for enhancements and bug fixes with the help of Ansible, bashscript and vim editor• Responsible for installation, configuration of Linux servers using jumpstart and interactive methods with Kubernetes base ideology• Used Ansible (YAML) playbooks in managing servers remotely and perform security hardening and maintenance on various systems including firewall and SELinux configuration• Configured DNS client on servers

• Provides technical leadership to a small team of engineers working on compiler middle-end optimizations and analyzed the performance of applications code running on GPUs with the aid of profiling tools• Identify opportunities for performance improvements in the LLVM based compiler optimizer and interact with open-source LLVM community to ensure strong and tighter integration• Design and develop new compiler passes and optimizations to produce best-in-class, robust, supportable compiler and tools• Worked with geographically distributed compiler, architecture and applications teams to oversee improvements and problems resolutions• Working closely with other engineers to build and manage infrastructure-as-code, cloud infrastructure, observability systems, and other mission critical systems and enjoy collaboration and partnering on hard problems to solve complex problems that arise at scale• Applied time-tested software principles to develop a network management platform that is user focused, intent-first, and built in layers using composable modules with clear schemas and single responsibility principle (Network Services)• Designing and configuring the network and monitoring & maintaining the network and providing specifications and detailed schematics for network architecture• Providing specific detailed information for hardware selection, implementation of techniques and tools for the most efficient solution to meet mission needs, including present and future capacity requirements• Designing and implementing failover solution for major customers and providing specific network solutions to support server requirements to include load-balancing, VPNs, firewall contexts, and network address translation (NAT) where appropriate• Understanding and solving business needs at scale with high-quality solutions and leaning into proactiveness and effective communication in pursuit of cross-functional alignment • Create and expand documentation

• Deployed and maintained Windows environment and Clusters on private and public cloud environments, hosting Enterprise and Engineering workloads• Installed, configured, tune and optimized Windows operating systems and applications and good understanding of Active Directory, DNS, DHCP and Group Policy and well-verse with Microsoft Failover Clustering• Performed regular system updates, and firmware upgrades and monitor system performance and ensure high levels of performance availability and security• Responsible for ensuring the stability, integrity and efficient operation of the production infrastructure server components and hands-on administration of Linux/Unix systems• Design of secure, robust, flexible and scalable systems and maintain an overall vision and direction for the infrastructure; and work in a team on global networking infrastructure• Design, configure and build secured systems, applying STIGs to Windows and Linux systems• Analyzed the requirements, design, and implement an effective system monitoring solution in Zabbix and Nagios• Developed project implementation documentation including all technical information needed for the successful implementation of the project and ensured that all components of the infrastructure are well documented and conform to standards• Provided On-call support to resolve OSS application issues after normal business hours• Follow instructions on various tasks including but not limited to building POAM’s, Data calls, and Network structure and directs junior colleagues effectively• Generate ISO files used for system installation with the appropriate patches• VMware management – setup , user account, network and datastore management• Build scale-out storage and data systems and maintained NAS system for NFS and SMB as well as supported internal and cloud systems• Open, track and closed trouble tickets as well as input trouble calls into ticket tracking system (Remedy, ServiceNow)

• Coordinated with internal departments to review existing integration capabilities, data sources and proposed solution designs for feasibility, cost and functionality; and conducted design sessions with appropriate participation from architects and engineers• Created detailed design documents and functional specifications for new applications, services and enhancements to existing systems and services• Installed, administered, configured and maintained check-In applications including operating systems and related software• Developed core code modules, unit test tools and release notes for enhancements and bug fixes with the help of ansible, javascript and vim editor• Reviewed new development tools, application frameworks and testing tools for functionality and effectiveness• Used established change management processes, requiring operational procedures be performed with minimal client impact• Integration of Linux systems into an Active Directory environment using SSSD/winbind. Also worked with database administrators to configure, tune and maintain databases in a variety of languages• Implemented a Windows-based VDI platform leveraging MS Hyper-V HCI Cluster, MS Storage Spaces, MS Azure and other supporting technologies. Also implemented and maintained MS SQL Server clustered environments to improve performance and resilience• Identified inefficiencies in established processes and developed PowerShell systems to automate solutions• Designed and implemented change and control policies and disaster recovery plans; led troubleshooting efforts to restore functionality in the event of an outage• Monitored and assisted in managing applications, device availability, network conditions and status, system reliability and performance, service and program maintenance and storage resources with Kubernetes• Build up and optimized server system benchmarks based on deep understanding of server system architect, key part performance matrix, and workload characterization

• Created and managed user/group accounts and setting password aging, permissions and account expiration• Used remote tools such as remote desktop, DameWare Utility to analyzed and resolved troubled issues• Configured hardware and software, installed Microsoft Windows Patches and updates, performed computer re-images, and assisted users with data backup and restoration• Active Directory user account, password and profile management on Windows Server usage on both Unix/Linux and windows-based servers• Monitored system performance and capacity planning in anticipation of system resource usage and needs; and ensured high levels of performance availability and security• Monitored system logs for unauthorized access, system errors, hardware failures and system health using native Linux tools • Proficient in utilizing SDKs and developing applications for various platforms• Skilled in building customized OS images and optimizing applications for performance enhancements using Ubuntu Core, desktop, and Server environments.• Infrastructure management: good understanding of virtualization technologies VMware, OpenStack, etc and experienced with configuration management tools like SaltStack/Ansible• Security and Compliance: Implemented and enforced security best practices and policies and ensured compliance with industry standards and regulations like CIS benchmarking• Troubleshooting and Support: Provided advanced technical support for Windows server-related issues. Collaborate with other IT teams to resolved complex issues, created and maintained detailed system documentation configurations and procedures

• Engineered and implemented a scalable global Microsoft Directory infrastructure that leverages technologies such as MS Active Directory and Domain Services• Enterprise Applications with Single Sign-On and Self Service Password Reset in Entra ID as well as Group Policy Management, Sites and Services in Active Directory and Intune endpoint management• Familiarity with advanced authentication systems such as privileged access management• Implemented a Windows-based VDI platform leveraging Microsoft Hyper-V HCI Cluster, MS Storage Spaces and other supporting technologies• Supported the maintenance of existing Microsoft platform infrastructure and Windows platform troubleshooting escalation requests• Worked with CPU, memory, SSD engineers, and vendors to setup the right performance target for the key parts for the server system. Solved performance related issues• Performed regular system updates and firmware upgrades• Monitored system performance and ensured high levels of performance availability and security• Based on a good understanding of mainstream applications, database, big data, storage, AI computing, profiling – worked with application team to update workload profiling system to solve performance problems and to achieve best performance on certain hardware• Co-work with server system architects to evaluate new architecture design• Co-work with cloud vendors, equipment vendors to evaluate the performance of cloud service or newly introduced hardware• Co-worked with CPU, memory, IO controller, SSD engineers and vendors to setup the right performance target for the key parts of the server system and solved performance related problems• Worked with application team to update work load profiling system and working together to achieve the best performance on certain hardware• Worked with industry consortiums and open standard committees to investigate emerging technologies or standards, and contribute research results and visions to the industry