About

I'm Mohammad Siddiqui, known to most people in the industry as Sarge. I'm a presales solution architect specialising in enterprise cloud infrastructure, based in Sydney, working across financial services, vendor, consulting and customer environments.

Most recently I was technical owner for multi-region VMware Cloud Foundation 9.1 programmes across Australia, Asia, the UK, Europe and North America, including large brownfield conversions from standalone vSphere to VCF-managed infrastructure. That work spans private, hybrid and public cloud, vSAN ESA, VKS, security and disaster recovery, and building multi-fleet resilience frameworks mapped to APRA CPS 234 and CPS 230, PRA SS1/21 and NYDFS.

My work sits where architecture meets the commercial decision. Running technical discovery, building the business case, designing the target state, and staying involved long enough to find out whether it held up.

That exposure is what this publication draws on. I have seen how a broad range of customers actually consume this technology: the patterns that work, the integrations that turn out harder than the datasheet suggests, and the design decisions that look fine until year two.

Where I am now

My role at Broadcom ended on 11 October 2026 when the position was made redundant. I am available immediately and talking to people about what comes next.

Presales and solution architecture is where I am most useful, because the job is as much about running a discovery honestly and surviving the business case questions as it is about the design itself. Enterprise platform architecture, migration and resilience, usually in regulated environments, and usually where the technical answer and the commercial one have to agree.

Sydney or the Gulf. I am open to relocating.

What I am working on while I have the time

Three things, and they turn out to be the same thing.

Kubernetes and GPU platform work. CKA first, then inference serving and GPU partitioning on real hardware rather than on paper. Most AI infrastructure lands on Kubernetes, which is also where a lot of virtual machine estates are heading, so it is the one layer where everything I have done and everything I want to do next actually meet.

Cloud to architect level across AWS and Azure. I have already moved VMware estates onto VMware Cloud on AWS, Azure VMware Solution and Google Cloud VMware Engine, so I know what each one costs and where each one breaks. The certification is to close the gap between having done it and being able to prove it.

And the data centre. Before VMware I ran data centres and built greenfield sites, and the thing said too quietly about AI is that it is a power, cooling and density problem before it is a model problem. The constraint has moved back into the building. That is not a new subject for me. It is an old one that came back.

The publication follows the same path, so what appears here over the next few months is the evidence rather than the intention.

If any of that is useful to you, sarge.siddiqui@icloud.com reaches me directly, or find me on LinkedIn. I answer everything that is not a bulk recruiter template.

vExpert

I have been a VMware vExpert for two years running, awarded under the Evangelist path. It gets mistaken for a certification fairly often, so it is worth saying plainly: there is no exam. The programme recognises what you contribute to the community around the technology, which in my case is this publication, my work with Sydney VMUG, and the ANZ lead role I held for APJ vCommunity until October 2026.

I have written up how the programme actually works, what it gets you, and what sits behind mine. That page also explains how to get involved with your local VMUG chapter, which is a more useful thing to do than collecting a badge.

Certifications and credentials

VMware vExpert
Awarded 2025 and 2026, Evangelist path. Community recognition rather than a technical certification.

VCP-VCF Architect
VMware Certified Professional, VMware Cloud Foundation Architect. Current to April 2029.

VCP-VCF Administrator
VMware Cloud Foundation administration. Current to July 2028.

VCAP6-CMA
VMware Certified Advanced Professional 6, Cloud Management and Automation Deployment. Awarded 2016.

VCP-DCV, continuously
VMware certified without a break since VCP5 in 2014, through VCP6 and VCP6.5 Data Center Virtualization and on to the current VCF track.

Microsoft Certified: Azure Fundamentals
Awarded 2023.

TOGAF
The Open Group Architecture Framework. Enterprise architecture method rather than a product certification, and the reason I tend to start with capability and constraint before technology.

In progress
Current-generation VCAP, on the way to VCDX. A longer road and a different kind of exercise entirely. CKA alongside it.

ANZ Technical Account Manager of the Year, 2017
VMware. An award rather than a credential, but it belongs in the same column.

Certification tells you someone sat an exam. It does not tell you whether they have run the thing at three in the morning. I list these because people ask, not because I think they settle anything.

What I work on

Most of my career has been in and around financial services. The VMware stack is where I go deepest, but very little of the work stays inside it, and the interesting problems are almost always at the boundaries.

VMware and Broadcom
VCF 9.x architecture and design. Fleet and multi-fleet topology. Workload domain design. Brownfield import and conversion. vSphere. vSAN and vSAN ESA. NSX and VPC networking. Avi load balancing. VCF Operations. VCF Automation and self-service. Identity Broker and SSO federation. vSphere Kubernetes Service. Tanzu. Data Services Manager. Cloud Director. Lifecycle and fleet patching. Certificate lifecycle. Disaster recovery and cyber recovery. Private AI on VCF. Memory tiering. Confidential computing.

AI infrastructure
GPU partitioning, MIG and vGPU. Inference serving and engine selection. KV cache and memory bandwidth behaviour. Quantisation trade-offs. Cost per token modelling. RAG as a distributed systems problem. GPU reference architecture and cabling. Agentic workflows and MCP.

Cloud and hyperscalers
Amazon EVS. Azure VMware Solution. Google Cloud VMware Engine. Oracle Cloud VMware Solution. Licence portability and BYOL. Hybrid and migration design.

Platform, network and hardware
Kubernetes. GitOps and Argo CD. EVPN and Cisco Nexus. Enterprise storage integration. Dell, HPE, NetApp, Pure. Supermicro HGX. ODM server economics. Bare metal lifecycle. Object storage.

Data centre
Greenfield build and site selection. Power, cooling and density. Data centre transformation and consolidation. Migration between facilities. The constraints that decide what you can actually run, before anyone opens a hypervisor.

Security, compliance and resilience
APRA CPS 230 and CPS 234. PRA SS1/21. NYDFS. STIG compliance. Post-quantum readiness. File integrity monitoring. Hypervisor-native EDR. Vulnerability and patch response. Resilience framework design.

Earlier ground, still useful
Backup and tape operations. Storage administration. Windows and database platforms. Enterprise networking. Media production systems. Technical account management.

The last group is there deliberately. A surprising amount of what I know about recovery design comes from having been the person who found out at 4am that the tape was blank.

How I got here

It started in a tape library. My first job was as a tape library and backup operator at the Commonwealth Bank, swapping cartridges and running overnight jobs. You learn things there that stay with you: that backups are only as good as the restore nobody has tested, and that the person who gets called at 3am has a different view of a design from the person who drew it.

From there I came up through storage and SAN, Windows desktop, server and database platforms, Active Directory, messaging, networking, JD Edwards and mainframe-backed systems, across HP, IBM, Cisco UCS and Dell.

At News Corp I was compute and VMware lead on a data centre transformation across a three-thousand-server estate, and the migrations I was running then were between buildings around Australia rather than to anyone's cloud. Before that I supported the media production systems behind the newsrooms, including on the sideline at NRL and cricket fixtures, working with photographers and journalists to get images and copy filed faster. Deadline pressure on a sideline teaches you something about latency that a datasheet doesn't.

I've sat on every side of the table: vendor at Compaq, HP, Dell and Lexmark, consulting at KPMG, and customer side at News Corp and the Commonwealth Bank.

I spent nearly a decade at VMware, most of it as a Technical Account Manager embedded with Australian banks: CBA, RBA, Westpac and Bankwest, in relationships measured in years rather than projects. Six years with RBA, six and a half with CBA, overlapping rather than one after another. Long enough to watch a design decision play out to its conclusion, which is a different education from making the call and moving on.

Six roles in twelve years at VMware and then Broadcom. Technical account manager, then senior, then staff, then sideways into presales and on to solution architect. VMware let me change lanes halfway through a career, which not every company would have.

So when I write about what enterprises get wrong, it isn't from the outside.

What you'll find here

Detailed technical write-ups on cloud infrastructure architecture and operations: design decisions and the trade-offs behind them, upgrade and patching strategy, identity and certificate integration, disaster recovery design, and third-party integrations in enterprise environments.

Not product news, not marketing. If something has a sharp edge, I'd rather say so.

New posts go out weekly to 410 subscribers. Free, and there is no paid tier.

There are over a hundred technical articles here now, which is too many to browse. Start here is a short list of the ones that best show what I do, grouped by the kind of problem they deal with.

Not everything here is technical. I also write about organisations that get things wrong, decisions that turned out to matter, and reflections on faith and practice under Deen. Same approach, different subject.

Getting in touch

Work and roles
sarge.siddiqui@icloud.com. Presales and solution architecture, enterprise platform work, AI infrastructure and data centre strategy. Sydney, the Gulf or Singapore. CV on request, and tell me what you are hiring for so I send the version that fits it. Also happy to talk about a specific design problem even if there is no role attached to it.

LinkedIn
linkedin.com/in/sargesiddiqui

Corrections
The fastest way to find a weak assumption is to have someone running the same thing in production tell you it does not hold. If I have got something wrong, or there is a design problem you would like covered, reply to any post or email me. I read everything.

Speaking and user groups
I am on the Sydney VMUG committee, and I was ANZ lead for APJ vCommunity until October 2026. If you want to speak at a chapter event, or you are thinking about it and are not sure, this page covers how to start and what we do to make a first session easy.

Views expressed here are my own. I no longer work for Broadcom, and everything published here is written from public documentation and my own experience rather than from any internal source.