Cloud Engineers Behind the Account Layout, the Network and the Move

Moves between cloud providers are turning into routine work. In a February 2026 press release, Gartner forecast that sovereign cloud spending will shift 20% of current workloads from global to local providers. Companies that hire cloud engineers now need someone who lays out accounts, networks and permissions so the next move stays a plan. Cortance matches engineers who have made such moves.

Decisions from the first month that come back in year three

Account structure, address ranges and identity rules are usually set while the product is small and one person owns the console. Nobody revisits them until a migration, an audit or a second region forces the question.

Two shortcuts cost the most to undo. When every environment shares one account, a single permission mistake reaches production, and nobody can say which team spends what. When a VPC uses the same address range as the office network or a partner, the two cannot be connected without readdressing one side, which means downtime.

Joining a corporate network to AWS

Vugar K. worked as a Cloud Network Engineer at Rapid7, a cybersecurity company. He connected corporate networks behind Cisco ASA and Check Point firewalls to AWS, and kept the Linux environments consistent with Chef and Puppet.

Hybrid links like that one are where undocumented firewall rules and overlapping address plans fail first. Vugar holds the AWS Certified Solutions Architect Professional and AZ-104 credentials.

Infrastructure for an energy clearing house

Elgun M. managed the infrastructure on Google Cloud and OpenShift for the risk development team at European Commodity Clearing, which clears trades for energy exchanges. He built and maintained the CI/CD pipelines and handled infrastructure requests from the developers, alongside his own Python and Angular work.

In a clearing house, an environment that nobody can rebuild is a risk on its own. Platforms like that one are where provider choice, regions and access rules get written down before workloads move.

Pricing a data-heavy workload before it is built

Imagery processing and AI workloads change the cost side of a cloud plan, because processing grows with data rather than with users. Viktoriia A. set up the processing environment for drone imagery used to monitor railway infrastructure, locally and on AWS. She benchmarked processing approaches and estimated the AWS infrastructure and cost needed to scale to production workloads.

That estimate is what turns a choice between a managed service and self-hosted processing into arithmetic. Ask a candidate for one they have made, and how close the bill came to it.

Allocat, moved to a second provider

Abrham M. took Allocat, a resource planning platform for IT agencies, from AWS to Google Cloud. Part of the platform moved through Terraform, and the rest moved by hand. Resources that were never written as code are the ones a move like this has to cover by hand.

Abrham's case is a fair test for any interview: ask which parts of a platform were in code before the move. Tell us your provider and the move you plan, and our matching returns profiles within 30 minutes; work can start within 2 days.