Core Skills for Working on Cloud Platforms and Services
By vaayaa Editorial Team · Subject panel: Hiroshi Tanaka (PhD, Computer Science), Carlos García (Master's, Data Science)
Cloud platforms and services are computing resources that providers deliver over a network, usually as infrastructure (IaaS), platform (PaaS) or software (SaaS). Working with them requires skills in core concepts, strategy, cost control, security, network access, and platform or application building, which the eight courses in this discourse cover.
Key takeaways
- Cloud services are usually grouped into infrastructure, platform and software models, and each model shifts work between provider and customer.
- Security responsibility is split, so providers secure the underlying facilities while customers secure their data, configurations and access.
- Cost control is an ongoing operating discipline, since survey respondents estimate that close to three in ten dollars of IaaS and PaaS spend is wasted.
- Network access controls such as VPN configuration are one part of a wider security design, not a complete solution.
- Platform engineering and SaaS development turn cloud features into tools that other people can use.
- Start with shared vocabulary, then pick the role-based path that fits your work.
Cloud platforms and services are computing resources that organizations rent over a network rather than buy and run themselves. The US National Institute of Standards and Technology describes cloud computing as on-demand network access to a shared pool of configurable resources that can be provisioned with little effort [1]. That short definition hides a large set of working skills.
These skills fall into a few groups: understanding the service models, planning and paying for cloud resources, securing them, and building platforms and software on top of them. Each group has its own tools and career paths, but they depend on each other. A cost decision affects security, and a security rule affects how a platform team ships software.
This discourse explains each group, compares the three largest skill areas, and offers a learning path, a worked example and a checklist. It is written for readers who may never take a course, and it separates sourced figures from editorial judgment.
Cloud Foundations and Infrastructure Models
Cloud services are usually grouped by how much of the technology stack the provider manages. The NIST definition names five essential characteristics: on-demand self-service, broad network access, resource pooling, rapid elasticity and measured service [1][2]. Each one changes how a team plans capacity and budgets.
Resource pooling means many customers draw on shared hardware, with resources assigned and reassigned according to demand. Rapid elasticity means capacity can scale outward and inward, in some cases automatically [3]. Together, these traits explain why a team can add servers in minutes rather than waiting weeks for purchase orders.
Basics of Cloud Computing Concepts builds this vocabulary from the ground up. It is a sensible first step for anyone who needs to read a cloud invoice, a proposal or an architecture diagram without guessing at the terms.
The three service models
- Infrastructure as a Service (IaaS): the provider supplies computing, storage and networking, and the customer configures the operating system and everything above it [2].
- Platform as a Service (PaaS): the provider manages the runtimes, libraries and operating systems, so developers focus on their code [7].
- Software as a Service (SaaS): the provider runs a finished application, and customers configure and use it rather than building it [2].
IaaS Infrastructure Management goes deeper into the first model. Learners study how virtual machines are provisioned, how networks and storage are organized, and how automated governance rules keep resources within policy. The work sits between a systems administrator and a cloud architect.
Opinion The NIST vocabulary is a practical starting point because job postings and vendor documentation often reuse its terms.
Spend management is the most widely cited challenge, and the estimated waste rate is higher for infrastructure and platform spend than for software.
Source: SHI Resource Hub [10]
View chart data
| Respondents naming cloud spend their top challenge | 85% |
| Large enterprises spending over 5 million dollars monthly | 76% |
| Estimated waste, IaaS and PaaS spend | 29% |
| Estimated waste, software spend | 25% |
Strategy, Consulting and Cost Control
Cloud consulting turns a technology choice into a business plan. A consultant checks whether an organization is ready to move, decides which applications should move first, and sets rules for governance. Cloud Consulting Services follows that sequence, from readiness assessment through migration and security governance. Compliance is shared as well, since control areas such as regulatory compliance belong to both provider and customer [5].
Cost is the most common pressure point. A Flexera survey, summarized by SHI, found that 85% of respondents see managing cloud spend as their top challenge [10]. The same summary reports that 76% of large enterprises spend more than $5 million a month on public cloud [10].
Cloud Cost Optimization addresses the waste side of that problem. Flexera's survey respondents estimated that 29% of their IaaS and PaaS spend is wasted, which is close to three dollars in every ten [9]. Flexera's analysis also notes that leading organizations are moving from cost management toward value management, which ties spending to what the business gains [11].
Common sources of waste
- Compute that runs at low utilization for long periods.
- Storage volumes that are no longer attached to any workload.
- Snapshots and backups kept beyond any retention need.
- Software licenses assigned to people who have left the organization.
Opinion Tagging every resource with an owner and a cost center is usually the cheapest first fix, because it makes waste visible to the people who can remove it.
Applying the survey waste estimate to larger budgets produces sums that grow in direct proportion to spend.
Source: SHI Resource Hub [10]
View chart data
| Monthly spend of 1 million dollars | 290,000 USD per month |
| Monthly spend of 5 million dollars | 1,450,000 USD per month |
| Monthly spend of 10 million dollars | 2,900,000 USD per month |
Securing Cloud Environments and Network Access
Cloud security begins with the shared responsibility model. Providers secure the facilities and underlying infrastructure, while customers secure what they place on that infrastructure [5]. CrowdStrike frames it as direct control: each party is fully responsible for the assets it controls, rather than the two sharing one asset [6].
The split changes with the service model. Under PaaS, the provider handles runtimes and operating systems, while the customer still owns application security, data security and user access [7]. Providers also draw their boundaries and tooling differently, so practitioners must read each provider's documentation closely [8].
Cloud Security Operations covers the operational side: designing controls, monitoring activity, managing identities and responding to events. Researchers note that some of the features that make cloud useful, such as self-service and measured use, can also open paths for abuse, so operations work has to account for them [4].
VPN Configuration focuses on one specific control: the virtual private network. A VPN creates an encrypted tunnel between a user or site and a private network, a pattern common when staff work remotely or offices connect to cloud resources. Configuring one well means choosing protocols, managing keys and certificates, and limiting access by role.
Opinion A VPN protects the path between two points, not the data once it reaches an application, so it works best alongside identity and access controls.
| Aspect | Cost management | Security operations | Platform engineering | Basis |
|---|---|---|---|---|
| Core question | Is each dollar of cloud spend producing value? | Who owns each control, and is it working? | How can teams deploy safely without manual steps? | Opinion |
| Typical tasks | Tagging, rightsizing, budget alerts, commitment planning | Identity management, logging, configuration checks, incident response | Building templates, self-service portals, deployment pipelines | Opinion |
| Common tools or methods | Cost reports, tagging policies, FinOps practices | Access policies, monitoring, encryption, VPN configuration | Infrastructure as code, container platforms, automated pipelines | Opinion |
| Evidence in this discourse | Survey estimate of 29% waste on IaaS and PaaS spend | Provider and industry guidance on shared responsibility | Published definitions of PaaS and SaaS | Fact |
| Typical roles | Cloud cost analyst, FinOps practitioner | Cloud security engineer, security operations analyst | Platform engineer, site reliability engineer | Opinion |
| Main risk if weak | Unused spend grows quietly each month | Misconfigured access or unmonitored activity | Teams build inconsistent or fragile environments | Opinion |
Building Platforms and Software on Cloud Services
Platform and software work sits on top of infrastructure. Platform as a Service lets internal developers request databases, runtimes and deployment pipelines through standard templates rather than through a ticket for each request. The aim is to cut repeated manual work, which platform teams often call toil.
PaaS Platform Engineering teaches how to design, operate and scale a private PaaS for an organization. The work covers self-service portals, standard templates and automated deployment. The goal is fewer manual steps and fewer errors when many teams ship software at the same time.
SaaS is the other side of the same stack. A SaaS provider runs one application for many customers [2]. SaaS Application Development covers multi-tenant design, in which a single codebase and shared infrastructure serve many separate customer accounts while keeping each account's data isolated.
Platform work compared with product work
- Platform teams build shared tools, templates and guardrails that other teams use.
- Product teams build features for end customers using those tools.
- Multi-tenant SaaS developers must plan isolation, billing and upgrades for every customer at once.
Comparing Three Core Skill Areas
The three largest skill groups in this theme are cost management, security operations and platform engineering. They overlap in daily work, but each answers a different question. The table below sets them side by side, with a basis label for every row.
Original analysis
vaayaa analysis: Estimated waste at three spending levels
- Start with the 29% estimated waste share for IaaS and PaaS spend reported in the survey [9][10].
- Apply that share to three monthly spend levels: 1 million, 5 million and 10 million dollars.
- Multiply each spend level by 0.29 to get estimated waste per month.
- Compare with the 25% software waste estimate [10]. At 1 million dollars, the 4 point gap equals 40,000 dollars per month.
Under these inputs, estimated monthly waste is 290,000, 1,450,000 and 2,900,000 dollars at the three spend levels. Small differences in percentage points become large sums at enterprise scale, which is why cost skills carry weight in planning.
A five-step learning path for cloud roles
- 1
Build vocabulary
Start with the basics of cloud computing concepts and the three service models.
- 2
Choose a role lens
Decide whether your next role centers on cost, security, platform building or consulting.
- 3
Add the core course
Take the infrastructure, platform or software course that matches that role.
- 4
Pair with a control
Add cost optimization or security operations so your technical work has an operating side.
- 5
Document one design
Write a short case note on one design, including its cost and security decisions.
Worked example
Estimating monthly waste for one team
- Assume the team spends 400,000 dollars per month on IaaS and PaaS (assumption for illustration only).
- Apply the 29% survey estimate: 400,000 x 0.29 = 116,000 dollars per month estimated as wasted.
- Assume the team recovers half of that waste through rightsizing and cleanup: 116,000 x 0.5 = 58,000 dollars per month.
- Annualize the recovered amount: 58,000 x 12 = 696,000 dollars per year.
Under these assumptions, the team could recover about 696,000 dollars a year. This is an illustration built from survey averages, not a forecast for any one organization.
Practical checklist before you choose a course
- Write down the job or project that the course should support.
- Confirm which service model your work uses: IaaS, PaaS or SaaS.
- List the tools and provider documentation you expect to read.
- Decide who owns security and cost tasks in your target environment.
- Pick one course to start, then one follow-up that adds an operating skill.
- Plan one small written case study to show what you learned.
Frequently asked questions
Do I need a technical background before starting these courses?
Basic comfort with networks, operating systems and files helps, especially for the infrastructure and VPN topics. The introductory course on cloud concepts is a good place to begin for readers from other fields. Business readers can start with cloud consulting or cost control, which rely more on planning and analysis than on coding.
What is the difference between cloud cost optimization and cloud consulting?
Cost optimization focuses on finding and reducing waste in existing environments. Cloud consulting covers a wider range of work, including readiness checks, strategy, migration planning and governance. A consultant may recommend cost changes, but the consulting role also covers decisions about what to move and how to govern it.
Who is responsible for security in a cloud environment?
Both the provider and the customer hold responsibilities. The provider generally secures the physical facilities and underlying infrastructure. The customer secures its data, identities, configurations and applications, and the exact split depends on the service model used. Reading each provider's documentation is necessary because boundaries differ.
Is a VPN enough to secure cloud access?
No. A VPN encrypts traffic between two points, which is useful for remote and site connections. It does not replace identity controls, least-privilege access, logging or monitoring. Most secure designs combine several controls, and a VPN is usually one layer among them.
How is a PaaS different from a private cloud?
A private cloud describes who uses the infrastructure, meaning a single organization. PaaS describes what the platform provides, namely runtimes, tools and services for building applications. A private PaaS is a platform built for one organization on infrastructure it controls or dedicates to its use.
Can these skills lead to a job without a degree?
Employers vary, and course completion alone does not guarantee a role. Many professionals combine short courses with work projects, certifications from providers and portfolio pieces. Written case studies that explain design, cost and security choices can help show what you have learned to a hiring manager.
Conclusion
Cloud platforms and services connect several skill groups: concepts, consulting, cost control, security, network access, platform engineering and SaaS development. The shared responsibility model and the service models give these skills a common frame, so each course reinforces the others. Survey figures show that cost and security are pressing concerns for many organizations, which makes them strong areas to learn.
Choose one role, learn the vocabulary, and then add one operational skill that protects what you build. Keep your own estimates separate from sourced figures, and treat every calculation as a planning tool rather than a promise.
References
- National Institute of Standards and Technology. The NIST Definition of Cloud Computing.
- National Institute of Standards and Technology. NIST Cloud Computing Reference Architecture.
- U.S. General Services Administration. Cloud Basics.
- arXiv. Considerations for Cloud Security Operations.
- ISACA. Understanding the Shared Responsibilities Model in Cloud Services.
- CrowdStrike. What is the Shared Responsibility Model?.
- Datadog. Simplifying the shared responsibility model: How to meet your cloud security obligations.
- Upwind. Understanding the Shared Responsibility Model in Cloud Security.
- Tech86. Cloud waste: 29% of cloud spend is wasted, says Flexera.
- SHI Resource Hub. FinOps takeaways: What Flexera's State of the Cloud Report tells us.
- Flexera. Cloud cost management isn't just about cutting waste.