Kubernetes enterprise adoption has already settled into place at the industry level. It is the default platform new applications get built on, and most infrastructure teams now assume they will be running it within a few years whether they chose it first or inherited it. What has not settled is the migration itself, specifically the move off VMware that Broadcom’s acquisition and licensing changes pushed onto a lot of roadmaps at once.
That migration is real, but it is slower and more labor-intensive than the pace most companies originally planned for. The technology path is not really in question anymore. What actually determines how fast a company gets there is whether it can find and keep the people qualified to run it.
That is the gap this article is about: wide adoption at the top, and a narrow, talent-constrained path to actually finishing the move.
The VMware migration is no longer the only thing pulling companies toward Kubernetes. Among organizations running generative AI in production, 66% already use Kubernetes to host some or all of their inference workloads, according to the same CNCF survey cited above. Portworx’s data points the same direction: 45% of its respondents already run, or plan to run, AI and machine learning workloads on Kubernetes.
What that means in practice is that a migration team is no longer being asked to handle one job. It used to be enough to understand container orchestration and move virtual machines over cleanly. Now that same team is also expected to support GPU scheduling, model serving, and the operational patterns that come with AI workloads, often on the same cluster. That raises the bar on who can actually lead this work, right as the talent pool for that work is already tight.
Kubernetes Enterprise Adoption & the VMware Migration
The migration funnel: wide adoption, narrow completion
Enterprise Kubernetes adoption is nearly universal. Finishing the move off VMware is a different story.
Run Kubernetes in production among companies already using containers
82%
Are actively reducing their VMware footprint
86%
Plan to proactively reduce their VMware footprint through Kubernetes specifically
69%
Have completed a full migration off VMware so far
4%
Why the funnel narrows: the talent gap
$148,750
US median salary for a DevOps Engineer, the role most migrations need to lead the work
75 days
Median time to fill a technical role in the US, rising further for senior hires
Sources: CNCF 2025 Annual Cloud Native Survey · CloudBolt, "The Mass Exodus That Never Was" (Jan. 2026) · Portworx, Voice of Kubernetes Report 2026 · Robert Half Salary Guide · Ashby Recruiting Operations Benchmarks
Use our free calculator to compare US hiring costs against nearshore bill rates for Kubernetes, DevOps, and platform engineering roles.
Migration budgets tend to get built around infrastructure costs: licensing, cloud spend, tooling. The people cost is usually treated as a rounding error, and that assumption does not hold up against current US salary data. A DevOps Engineer in the US now commands a national salary range of $119,500 to $177,250, with a midpoint around $148,750, according to Robert Half’s current salary data. A Site Reliability Engineer, the role most directly responsible for keeping a Kubernetes environment stable in production, runs $116,750 to $175,500.
Those numbers are national medians, not outliers in expensive metros, and they apply before accounting for the premium that specific Kubernetes certifications or AI infrastructure experience tend to add on top. A migration that needs two or three people at this level, on top of whatever the rest of the infrastructure team already costs, can add six figures a year to a budget that was built around licensing math rather than headcount. Latin American nearshore consultants in equivalent Kubernetes, DevOps, and platform engineering roles typically run 40 to 60% below those US figures, a gap Fast Dolphin’s own senior placement data lays out in detail in the Nearshore Cost Advantage Report.
Even with the budget approved, the clock is the bigger problem. Time-to-fill for tech roles nationally reached 51 days in February 2025, up from 48 days a year earlier, which iCIMS notes is 10 days longer than the overall labor market. That is the average across all tech roles, not the senior, specialized hires a migration actually needs.
Specialized roles take longer than that average suggests. Ashby’s 2026 recruiting benchmark data, drawn from more than 54 million applications and 93,000 jobs, puts the median time to first fill at 75 days for technical roles against 60 days for business roles, and senior positions take 37% longer to fill than junior ones. Once sourcing, interview loops, and a standard notice period are factored in, a senior Kubernetes or platform engineering hire landing somewhere between 60 and 90 days is not a worst case. It is closer to the realistic middle.
For a migration running against a fixed VMware renewal date, that timeline is the whole problem. Fast Dolphin delivers a first candidate in 24 to 48 hours rather than starting a search from zero, which is the gap that nearshore IT staffing and headhunting for senior engineers is built to close.
It would be one thing if the company’s existing platform team simply absorbed this work. In practice, that team is usually already short-staffed. The Linux Foundation’s 2025 State of Tech Talent report found that 59% of organizations are understaffed in cloud computing and 56% in platform engineering specifically, figures drawn from a survey of hiring and training leaders across the industry.
Asking that same understaffed team to run daily operations and lead a multi-year VMware migration at the same time is how both jobs end up done worse. One response is to treat the migration as its own project with its own dedicated capacity rather than folding it into an already full workload, which is the logic behind bringing in temporary tech staffing to scale IT and engineering teams for the length of a defined initiative rather than permanently growing headcount.
Kubernetes migrations and the operations that follow them are frequently incident-driven. A cutover window slips, a pod starts crash-looping in production, or a cluster upgrade needs a decision made within the hour, not by the next morning. CNCF’s own survey data reflects this kind of operational friction indirectly: for the first time, organizational culture and cross-team coordination, rather than technical complexity, ranked as the top barrier to cloud native adoption.
Staffing models built around distant time zones, common with teams based many hours away in Asia or Eastern Europe, push that kind of coordination into async handoffs by default. A production issue that surfaces at 2pm Eastern waits until the next business day on the other side of the world. Latin American teams typically work within one to three hours of US business hours, which means the same issue gets a same-day response instead of an overnight queue. That overlap is also why nearshore time zone overlap tends to matter more for Kubernetes and DevOps roles than for project work that can tolerate a wider gap.
A narrow, container-only hiring search is rarely enough anymore. The same Linux Foundation research cited earlier found that 68% of organizations lack enough AI and machine learning skilled employees, and that shortage compounds with the cloud and platform engineering gaps already described rather than sitting apart from them. A Kubernetes migration that also needs to support AI workloads is drawing from three overlapping, equally scarce talent pools at once.
That compounding effect is part of why companies increasingly look past a single domestic hire and toward a wider pool of nearshore specialists, including generative AI developers in Latin America who already work inside Kubernetes-based environments rather than treating AI infrastructure as a separate specialty to source later.
Put the pieces together and a pattern shows up. A migration already running against a fixed VMware renewal deadline is being led by roles that are both scarce and expensive in the US, staffed by a team that was already stretched before the project started, and unable to lean on staffing models built around distant time zones for work that depends on real-time response. None of these pressures is new on its own. What is new is how often they are now hitting the same project at the same time.
The fix is not a better Kubernetes distribution or a more detailed migration runbook. It is rethinking where the specialized people running that migration actually come from. A company that keeps restarting a US-only search every time a role opens is solving a staffing problem with a technology plan, and that is why the gap tends to widen instead of closing on its own.
Each of the pressures above has a direct answer. Cost pressure gets addressed through nearshore rates benchmarked against US hiring rather than a vague promise of savings. The 60-to-90-day hiring timeline gets addressed with a first candidate delivered in 24 to 48 hours, not another open requisition. A stretched internal team gets relief through temporary staffing built around standing up a dedicated migration team without pulling existing platform engineers off their daily responsibilities. Time zone friction gets addressed by working with professionals based one to three hours from US business hours rather than on the other side of the globe. And the skills gap itself gets addressed through a vetted pool of nearshore Kubernetes, cloud, and AI engineers who already work inside the same stack a migration depends on.
Once a migration lead or platform engineer has proven out over a project, converting that person to a permanent member of the team is its own decision, and that is where Direct Hire Staffing fits. Every Direct Hire placement includes a 90-day replacement guarantee: if the professional leaves or is let go for documented poor performance within the first 90 days, the replacement is at no cost, or the fee is refunded on a pro rata basis if a replacement is not possible. The engagement model should match where the project actually is, not force a long-term commitment before anyone knows if the team is right.
Nearshoring and offshoring both move hiring outside the US, but the similarities mostly end there. Fast Dolphin’s breakdown of nearshore vs offshore vs onshore models walks through where the two approaches diverge on cost, time zone overlap, and day-to-day collaboration, and the short version is that nearshore teams in Latin America typically overlap six to eight hours with a US workday, while offshore teams in Asia often overlap two hours or none at all.
Cost still favors hiring outside the US, and that advantage holds for both models on paper. The real difference shows up in how a distributed team actually functions. Standups, code review, incident response, and the dozens of small handoffs that make up a normal sprint all get harder the less the workday overlaps, and a detailed cost comparison between US and Latin American development teams shows the savings gap holds even before you factor the time zone difference into total cost of ownership.
Brazil and Mexico alone are home to more than 2.2 million software engineering professionals, and together the two countries graduate more than 350,000 new engineering students every year, according to Deloitte’s research on software engineering nearshoring. That’s a talent base that’s grown fast enough to support specialized hiring across cloud, data, DevOps, and ERP, not just general application development.
Fast Dolphin’s own look at the best countries for nearshore software development breaks down how Mexico, Brazil, Colombia, Argentina, and Chile each stack up on cost, specialization, and talent depth, and a separate piece digs into why Mexico has become a top nearshore IT destination for companies that want the closest possible time zone alignment with the US.
Colombia has built a reputation of its own, particularly for mid-level developers and QA talent, even though its overall workforce is smaller than Brazil’s or Mexico’s. A dedicated guide to hiring software engineers in Colombia covers what US companies should know before staffing there. For most US companies, the right country isn’t a single default. It depends on the specific role, the budget, and how much time zone overlap actually matters for that team.
For HR and Talent Acquisition leaders staffing their own IT and Engineering teams, a few questions are worth asking before the next senior requisition goes out. Does the role genuinely need to sit in the US? Does the internal team have the bandwidth and expertise to source and vet candidates outside the country on top of everything else on their plate? Would a trial period attached to the hire reduce the risk of a bad match before committing to a permanent one?
Those questions tend to point in the same direction: toward a partner that already has the sourcing infrastructure, the vetting process, and the time zone alignment in place, rather than trying to build that capability from scratch inside a recruiting team that’s already stretched across every other open role.
Fast Dolphin has placed IT and Engineering professionals from Latin America with US companies for more than 20 years, and that focus is deliberate. We run exactly two staffing models, temporary staffing and direct hire staffing, rather than spreading attention across every kind of hire and every market. Temporary Staffing includes Contract-to-Hire as a built-in option, so you can start a placement on a trial basis with a clear path to bringing the professional on permanently if it’s the right fit, without treating it as a separate service to negotiate later.
We can have your first vetted candidate in front of you in 24 to 48 hours, at a cost significantly lower than US-based hiring, and every consultant goes through a vetting process built specifically for technical roles rather than a general screening process applied across every kind of hire. When something comes up, you’re working with people who know your account directly, not a support queue.
For Direct Hire Staffing placements specifically, every hire includes a 90-day replacement guarantee: if the professional leaves or is let go for documented poor performance within the first 90 days, we replace them at no cost, or refund the fee on a pro rata basis if we can’t.
Tell us about the roles slowing your VMware-to-Kubernetes migration down, and a member of the Fast Dolphin team will follow up within one business day.
The adoption data backs it up. Among companies already running containers, 82% run Kubernetes in production, up from 66% two years earlier, according to the Cloud Native Computing Foundation’s annual survey. Kubernetes enterprise adoption has moved from an emerging choice to the assumed baseline most infrastructure teams plan around.
Longer than most initial plans assume. Gartner’s guidance, based on research presented at its IT Symposium, puts a full migration at three years or more, and current survey data shows only 4% of companies actively moving away from VMware have completed that move so far. Treating it as a multi-year program rather than a single cutover tends to produce better outcomes.
Time-to-fill for tech roles nationally sits around 51 days on average, and benchmark data for technical roles specifically runs closer to 75 days, with senior roles taking 37% longer than junior ones. Combined with a documented shortage of platform engineering and cloud talent, a 60-to-90-day search for a senior hire is common rather than unusual.
US DevOps Engineer salaries currently range from about $119,500 to $177,250 nationally, with Site Reliability Engineers running a similar range. Latin American nearshore consultants in equivalent roles typically bill at 40 to 60% below those figures, without the handoff delays that come from staffing models built around distant time zones.
Time zone overlap is the key difference between nearshore and more distant staffing models. Latin American professionals typically work within one to three hours of US business hours, which supports same-day incident response and live collaboration during a migration, rather than the overnight async pattern common with teams based many time zones away.
Temporary staffing fits a migration with a defined scope and timeline, since it adds capacity without a permanent headcount commitment. Direct hire staffing fits better once a role proves out and the company wants that person leading Kubernetes operations on a permanent basis, backed by a 90-day replacement guarantee if the placement does not work out.
