NVIDIA-Certified Professional: AI Rack and Interconnect
Issued Oct 2026 · valid to Oct 2028
Verify on NVIDIAHPC · AI Factory · Platform architecture Singapore
Amitabh
Pandey.
I work on the infrastructure behind high-performance computing and modern AI: GPU compute, high-speed fabrics and parallel storage, built on more than 23 years of Linux, cloud and enterprise operations.
HPC
Supercomputing · University of Edinburgh & PRACE
5× NVIDIA
AI infrastructure certifications, active to 2028
BeeGFS·WEKA
Certified on HPC parallel file systems
23+
Years in infrastructure
01 Certifications
Certified where AI infrastructure is built.
Exam-based certifications first, ordered by relevance to AI factories today. NVIDIA credentials link to their issuer records; other digital badges link to Credly.
Tier 1
AI factory, GPU cloud & HPC
Current focusNVIDIA-Certified Professional: AI Infrastructure
Issued Mar 2026 · valid to Mar 2028
Verify on NVIDIANVIDIA-Certified Associate: AI Infrastructure and Operations
Issued Jan 2026 · valid to Jan 2028
Verify on NVIDIAHPC storage & cluster management
ThinkParQ
BeeGFS
Parallel file system built for HPC and AI workloads.
- Certified System EngineerOct 2024
WEKA
WEKA Data Platform
High-performance data platform for AI and HPC storage.
- Technical XpertCertified
- Technical XplorerCertified
Bright Computing · now NVIDIA
Bright Cluster Manager
HPC cluster provisioning and management, now NVIDIA Base Command Manager.
- Basic AdminAug 2021
- OpenStackAug 2021
- Data ScienceAug 2021
- Bright 8.0 AdminApr 2020
Tier 2
Cloud & cloud native
Earned 2020 – 2021 · exam-based
Tier 3
Enterprise foundation
Linux · virtualization · storage · delivery
Red Hat Certified System Administrator
Linux systems administration on Red Hat Enterprise Linux.
Red HatCertificate
Jul 2012VMware Certified Professional on vSphere 5
VCP-510: enterprise virtualization on vSphere.
VMwareCertificate
VMware Certified Professional – Cloud
VCP-Cloud: private cloud infrastructure on VMware.
VMwareCertificate
Jul 2012NetApp Certified Data Administrator
Data ONTAP 7-Mode storage administration.
NetAppVerify
Citrix Certified Administrator for XenServer 6
Server virtualization on Citrix XenServer.
CitrixCertificate
Jun 2014PRINCE2® Registered Practitioner
Project management, with PRINCE2® Foundation.
PeopleCertCertificate
Dec 2007ITIL Foundation v3
IT service management fundamentals.
EXINCertificate
EXIN Cloud Computing Foundation
Cloud computing concepts and service models.
EXINCertificate
Course completions from the Linux Foundation, NVIDIA and universities are listed separately under Continuous learning.
Full Credly wallet
02 Profile
Behind the infrastructure.
My professional path has followed the evolution of enterprise computing: from Linux servers, network engineering and operational resilience to virtualization, cloud systems, high-performance computing and, increasingly, AI infrastructure.
I'm interested in the engineering layers that make demanding compute platforms dependable: hardware, networks, storage, orchestration, observability, and the processes that keep services running. The common thread is a systems perspective, designing and operating the foundations that applications and compute workloads depend on.
From enterprise Linux and networking to AI factories and GPU-powered computing platforms.
- Focus
- AI Factory · HPCPlatform architecture & engineering leadership
- Experience
- 23+ yearsEnterprise infrastructure to AI platforms
- Certified
- NVIDIA · BeeGFS · WEKA5 NVIDIA AI infrastructure certifications, plus HPC parallel storage
- Studied
- SupercomputingThe University of Edinburgh & PRACE, 2018
- Recognised
- J.P. MorganOutstanding Performance & Service Milestone awards
- Location
- SingaporeCareer across Singapore, Bengaluru & Greater Delhi
03 Capabilities
The stack, layer by layer.
Every layer of an AI factory, from the platform down to the operating system, and the certifications that back each one. The highlighted layers are the current focus.
↑ Workloads & platformskills · backed by
L7
AI platform & economics
Running GPU capacity as a cloud service and measuring the value of the tokens it produces.
GPU cloud platformsKubernetes cluster lifecycleAI tokenomicsKubeflow
Backed by
L6
AI compute & operations
GPU clusters for training and inference, and the cluster management that keeps them running.
Core focusGPU infrastructureRack & interconnect deploymentAI cluster operationsBase Command ManagerBright Cluster ManagerHPC
L5
AI networking
High-bandwidth fabrics that connect GPUs, on a long record in routing, segmentation and failover.
Core focusMellanox InfiniBandAI fabricsRouting & VLANFirewalls & VPNWAN failover
Backed by
NCP-AIN
2006 router project
L4
Storage & parallel file systems
Feeding data to GPUs fast enough, from parallel file systems to enterprise arrays.
Core focusBeeGFSWEKASpectrum Scale (GPFS)NetApp ONTAPHPE DMFZFS
Backed by
L3
Cloud native & automation
Containers, orchestration and infrastructure as code, on public cloud and on premises.
KubernetesAWSTerraformArgoTektonContainers
Backed by
L2
Virtualization & Linux
The operating layer underneath it all: Linux, Unix and enterprise virtualization.
RHELLinux / FreeBSDSolarisVMware vSpherevCloud & SRMCitrix XenServer
Backed by
RHCSAVCP-510VCP-CloudCitrix CCA
L1
Delivery & service management
Running infrastructure as a service: change, continuity, coordination and leadership.
PRINCE2ITILPatching & changeProject coordinationSix Sigma Green Belt course
Backed by
PRINCE2 PractitionerITIL v3
↓ Bare metal & data centre23+ years deep
04 Journey
Through every generation of infrastructure.
Each stage added a layer, from Linux and network resilience to enterprise operations, cloud and Kubernetes, and now AI factories. Full role history on LinkedIn
-
Now
AI Factory & HPC infrastructure
Platform architecture and leadership at an AI infrastructure company (associated with Firmus Technologies).
-
2026
NVIDIA AI infrastructure track
Earned NVIDIA certifications in AI Networking, AI Infrastructure & Operations, AI Operations and AI Infrastructure from January to March. AI Tokenomics Fundamentals followed in September, then NVIDIA-Certified Professional: AI Rack and Interconnect in October.
NCP-AINNCA-AIIONCP-AIONCP-AIINCP-ARI -
2024–25
GPU cloud & parallel storage
Certified BeeGFS System Engineer (2024) and Rafay Certified GPU Cloud Professional (2025), with NVIDIA Base Command Manager, Kubeflow, Argo and Tekton training.
-
2020–21
Cloud & Kubernetes
Five AWS certifications in three months, including Solutions Architect and DevOps Engineer Professional, then CKA, CKAD, Terraform Associate and Bright Cluster Manager.
-
2018
Supercomputing study
Supercomputing with The University of Edinburgh & PRACE, and High Performance Computing in the Cloud with Dublin City University.
-
2012–14
Virtualization, storage & delivery
VMware Certified Professional on vSphere 5 and NetApp NCDA in 2012; PRINCE2 Foundation and Registered Practitioner in 2014.
-
2007–09
J.P. Morgan · Bengaluru
Service Milestone Award (2007) and an Outstanding Performance Award for DST server patching (2009). ITIL Foundation v3 (2007).
-
2006
Mercantila Software Services · Bangalore
Built an open-source Linux router and firewall with ISP failover. Best Performer award.
05 Selected work
Systems that stay up.
Completed · Mar – May 2006 · Mercantila Software Services
Open-source Linux router & firewall
A Linux-based routing and firewall platform bringing monitoring, segmentation, traffic control, intrusion detection and ISP failover into a single network design.
LinuxFreeBSDiptablesOpenVPNSnortSquid
- Four NICs: one LAN, two WAN, one DMZ
- Security policy and routing with iptables
- DMZ and VLAN segmentation of Linux servers
- Site connectivity over an OpenVPN tunnel
- Snort IDS, Squid and class-based queuing
- Automatic failover to a secondary ISP
AI Factory & HPC infrastructure
The compute, networking, storage and operating practices underpinning modern AI factories and high-performance computing systems.
AI factoryHPCGPUInfiniBand
Enterprise infrastructure operations
Infrastructure operations, virtualization, patching and service delivery, including recognition from J.P. Morgan and specialist vendor training.
VirtualizationPatchingITIL
06 Recognition
Recognition through delivery.
Awards and appreciation earned for operational work, leadership and delivering on time.
J.P. Morgan2009
Outstanding Performance Award
For outstanding performance on DST server patching, Bangalore.
J.P. Morgan2007
Service Milestone Award
Marking three years of service.
Mercantila2006
Best Performer
Mercantila Software Services, Bangalore.
J.P. MorganLeadership
Excel Leadership Award
Recognition from the J.P. Morgan Excel leadership programme.
J.P. MorganAppreciation
Letter of appreciation
From the Global Co-CEO / CIO (Investment Banking) and GTI.
Fujitsu AsiaService
Long Service Award
For completing five years of service.
12.
Recommendations from colleagues and collaborators on LinkedIn, across technical infrastructure and client-facing delivery.
Read them on LinkedInHighlights his AI infrastructure, HPC, AI Factory and GPU computing work, across bare-metal and cloud, and his problem-solving.
Highlights how he handles difficult clients, and his consulting-style communication.
07 Continuous learning
Courses, kept apart from the exams.
Course completions and vendor training. They are listed here for completeness and not counted as certifications above.
Linux Foundation 2021–25
- DevOps & Workflow Management with Argo
LFS256 - CI/CD with Tekton
LFS274 - Intro to AI/ML Toolkits with Kubeflow
LFS147 - Kubernetes Security Essentials
LFS260 - Containers Fundamentals
LFS253 - Kubernetes for Developers
LFD259 - Kubernetes Fundamentals
LFS258 - Fundamentals of Quantum Computing
LFQ101 - Quantum Computing for Senior Leaders
LFQ102
NVIDIA 2020–24
- Base Command Manager Administration
2024 - Introduction to Base Command Manager
2024 - NVIDIA AI Enterprise Administration
v3.0 - AI Infrastructure & Operations Fundamentals
- Introduction to AI in the Data Center
2020
HPC & cloud 2018+
- Supercomputing
Edinburgh & PRACE - HPC in the Cloud
DCU - HPC with Containers: Singularity & Docker
- IBM Cloud: Microservices with Kubernetes
Coursera - LXC/LXD Deep Dive
A Cloud Guru - Python 3 for System Administrators
A Cloud Guru
Storage & clustering 07
- IBM Spectrum Scale 4.1 (GPFS)
H005G - IBM Platform Symphony
H050G - HPE DMF Administration
H6LG1S - Oracle Solaris 10 ZFS
SA-2290 - Red Hat Clustering & Storage
RH436 - Red Hat Enterprise Virtualization
RH318 - Veritas Cluster Server
5.0
VMware & Solaris 07
- vSphere: Install, Configure, Manage
v5.0 - vSphere: Troubleshooting Workshop
v5.1 - vCloud Director Fundamentals
v5.1 - vCenter Site Recovery Manager
v5.1 - Solaris Containers
SA-355 - Solaris 10 Logical Domains
SA-345 - Opsware Server Automation System
Management & data 05
- Six Sigma Green Belt
- Excel Leadership Training
JPMC - ITIL Foundation v3
- PRINCE2 Foundation & Practitioner
- Cloudera Administrator for Apache Hadoop
08 Get in touch
Let’s talk AI infrastructure.
I'm interested in the platforms and engineering practices behind large-scale AI systems, especially compute, networking, storage, cloud and high-performance infrastructure. For professional conversations, connect with me on LinkedIn.