How we work: from diagnosis to continuous evolution

What is WIP’s working method?

The method has five stages: diagnosis, plan, execution, operation and evolution. Not every client goes through all of them: a project may end at execution, and an ongoing contract lives in the last two. The order, however, never changes: nothing is executed before it is understood.

1. Diagnosis

Understand the system, the business and the problem before proposing anything. We measure what exists: architecture, code, infrastructure, backups, security and costs. The result is a snapshot with prioritized risks.

2. Plan

Turn the diagnosis into decisions: what to do, in what order, at what cost and with which acceptance criteria. The plan also states what is left out and why.

3. Execution

Do what was agreed, in small, tested changes that can be rolled back. Each relevant step is measured before and after.

4. Operation

Keep the system healthy in production: monitoring, alerts, tested backups, updates and incident response, within the contracted scope.

5. Evolution

Improve continuously based on what operations show: cost and performance trends, incidents and new business needs. Evolution feeds a new cycle of diagnosis and planning.

What principles guide technical decisions?

Decisions follow six well-architected design principles, which apply to any cloud provider or to your own servers.

Principle	The question it answers
Operational excellence	Can the system be released, observed and fixed safely, without relying on heroics?
Security	Are data and access protected, with least privilege and secrets kept out of the code?
Reliability	Does the system recover from failures, and has that recovery been tested?
Performance efficiency	Are resources used in line with demand, with no waste and no bottlenecks?
Cost optimization	Does every dollar spent on infrastructure have a reason, an owner and ongoing tracking?
Sustainability	Does the system avoid unnecessary use of computing resources and energy?

The principles aren’t boxes to check: they’re criteria for choosing between options. When two of them conflict, such as redundancy (reliability) versus cost, the trade-off is explained and decided together with the client.

The From Code to Production checklist puts these principles into concrete form: verifiable items, organized into seven pillars, that guide WIP’s diagnoses and that anyone can consult.

How does WIP use artificial intelligence?

WIP uses AI to be more productive in assisted coding, automated analysis, testing and documentation. Critical decisions, security and validation stay with designated owners: nothing produced by AI goes to production without an engineering review.

• Where AI helps: writing and reviewing repetitive code, analyzing logs and configurations, generating tests, drafting documentation and speeding up investigations.
• Where AI doesn’t decide: architecture, production changes, security, access to client data and any recommendation involving significant cost or risk.
• How validation works: whatever AI produces goes through the same path as any other change, with review, testing and a written record.

Using AI tools with client code or data is agreed at the start of each project, within the confidentiality terms in place.

What are the commercial terms?

The commercial terms exist so the client knows, before anything starts, what they are paying for, to whom and why.

1. Infrastructure separate from fees. Infrastructure, licenses, cloud usage and third-party services are listed separately from the amount paid to WIP.
2. Defined scope. Every project and every plan defines environments, hours or work capacity, support hours and exclusions.
3. Price before work. Published prices are reference values; the final scope and price are set in the proposal, before work begins. Services that depend on a diagnosis are priced after it.
4. No promise of absolute availability, nor of guaranteed recovery without a validated architecture and tests to back the promise. Availability and recovery targets, when they exist, are defined in the contract.
5. Scope changes agreed in advance. If something new comes up along the way, it becomes a separate proposal, approved before any work is done.

How does WIP communicate with clients?

Communication is in writing, in plain language, and always answers three questions: what is being done, what it costs and what the risks are. No one needs to master technical terms to follow the work.

• Two-layer reports: a summary for decision makers and the technical detail for the people doing the work.
• Recommendations with cost and reasoning: every suggestion says what it solves, what it costs and what the risk is of not doing it.
• Problems flagged early: if something goes off plan, the client knows before getting the bill or feeling the impact.
• Decisions on record: what was decided, by whom and why, for future reference.

Channels and update frequency are agreed at the start of each project or contract.