Infrastructure Strategy and Technology Selection Questions
Setting technical direction for infrastructure and deciding what to build on. Covers infrastructure vision and long-term roadmap, modernization and technical-debt strategy, platform decisions, and organizational and governance considerations, alongside the decision framework for adopting or retiring technology: build-versus-buy-versus-cloud-versus-on-premises trade-offs, vendor and platform evaluation, technology-portfolio rationalization, requirements-driven service selection, and weighing total cost of ownership and risk. It also covers managing an existing vendor relationship: quantifying and mitigating lock-in, negotiating contract terms, and planning an exit from a service the organization depends on. The leadership-altitude discipline of directing an infrastructure estate over time, building consensus among stakeholders with conflicting priorities, and persuading leadership or a team to accept a major technology change.
When evaluating a managed platform vendor, what security and compliance questions should you make sure to ask? Cover encryption and key management, identity and access controls, audit logging, data residency, breach notification, and relevant certifications.
Sample Answer
Direct answer
Ask the vendor to show evidence, not describe intentions, across six areas: encryption and key management, identity and access controls, audit logging, data residency, breach notification, and certifications. A vendor that answers fluently but cannot point to a document, a report, or a contract clause for each one has not actually answered.
Structured elaboration
| Area | What to ask | Why it matters |
|---|---|---|
| Encryption and key management | Is data encrypted at rest and in transit by default? Can we bring or manage our own keys through a key management system (a service that generates, stores, and rotates encryption keys), and can we revoke a key to make our data unreadable to the vendor? | Default-on encryption is table stakes; customer-managed keys are the difference between "the vendor promises not to look" and "the vendor is technically unable to." |
| Identity and access controls | Do you support single sign-on and short-lived federated credentials (OIDC, OpenID Connect, a standard for proving identity across systems without sharing a password) rather than only long-lived API keys? Is access role-based and can we scope a role to least privilege? | Long-lived shared credentials are the most common root cause of a third-party breach; federation lets you revoke access centrally the moment an employee leaves. |
| Audit logging | Do you log administrative and data-access actions, how long are logs retained, and can we export them to our own SIEM (security information and event management system)? | If the vendor is breached, your own retained logs are often the only way to determine what of yours was touched, independent of what the vendor tells you. |
| Data residency | Which physical region(s) does our data (including backups) live in, and can we pin it? | A verbal "we're global" answer often means backups silently replicate to a region your regulator does not permit. |
| Breach notification | What is the contractual notification window, and does it match or beat the legal requirement you are subject to (for example, the General Data Protection Regulation's 72-hour notification requirement to the relevant authority)? | A vendor's marketing page and its actual signed contract can promise different numbers; only the contract is enforceable. |
| Certifications | Which certifications (SOC 2 Type II, an independent auditor's report on a vendor's security controls; ISO/IEC 27001; and for payment data PCI DSS) do you hold, and does the audit's stated scope actually cover the specific product or service we are buying? | Vendors with a large portfolio sometimes hold a certification for one product line and imply it covers all of them. |
Worked example
Evaluating a payments-adjacent vendor for a fintech company serving European Union customers: you ask for the SOC 2 Type II report and find its scope explicitly excludes the newer API product you actually intend to use, which is the real finding, not "do they have SOC 2." You ask about breach notification and the contract says 30 days; your own regulatory obligation under GDPR (the EU's General Data Protection Regulation) is to notify the supervisory authority within 72 hours of becoming aware, so a 30-day vendor clause would make you non-compliant before you even know there was an incident, and that gap has to be renegotiated or the vendor rejected. You ask about key management and learn keys are vendor-managed only, meaning a subpoena or an internal vendor incident gives them technical ability to read your data regardless of their policy promises.
Trade-offs and pitfalls
Treating a certification badge as the finding instead of reading the report's actual scope is the most common mistake; a badge tells you an audit happened, not that the product you're buying was inside it. Pre-sales teams will often answer these questions optimistically without engineering sign-off, so ask for the security or compliance team directly rather than accepting an account manager's paraphrase. Finally, resist writing a purely defensive checklist with no way to fail a vendor: decide in advance which answers are disqualifying (no customer-managed keys for regulated data, no contractual notification window at or better than your legal obligation) so the exercise produces a decision, not just a document.
A client must meet strict regulatory requirements (for example PCI-DSS and GDPR). How would you incorporate that security and compliance work into a multi-year infrastructure roadmap - technical controls, process changes, milestones, and the evidence you'd prepare for audits?
Sample Answer
Regulatory requirements like PCI-DSS (Payment Card Industry Data Security Standard, the security standard payment networks require of anyone storing, processing, or transmitting cardholder data) and GDPR (General Data Protection Regulation, the EU's data-protection law) belong woven into the roadmap from the start as concrete technical controls, process changes, and milestones with evidence built in as a byproduct of normal operation, not as a phase bolted onto the end or an evidence-gathering fire drill right before the audit.
Technical controls
Map each regulation's actual requirements to specific architecture decisions rather than treating "compliance" as one undifferentiated bucket of work. PCI-DSS requires network segmentation isolating the cardholder data environment from the rest of the system, encryption of cardholder data both in transit and at rest, and strict, auditable access logging. GDPR requires supporting data-subject rights (the ability for an individual to request access to or erasure of their personal data), tracking a lawful basis for processing that data, and, often, safeguards around transferring EU personal data outside the EU.
Process changes
A recurring access-review cadence (confirming who still needs access to regulated systems), a formal change-management process with segregation of duties (the person deploying a change to a regulated system is not the same person approving it), and incident-response runbooks that account for regulator notification timelines, GDPR's 72-hour breach-notification window is a specific, real deadline worth naming and building the runbook's timeline around.
Milestones
Sequence controls to land before the workload that needs them goes live, not as a remediation sprint after a security assessment finds the gap. Build in a pre-audit readiness assessment milestone well ahead of the actual audit window, so findings get fixed on the team's own schedule rather than the auditor's.
Evidence for audits
Design systems so that audit evidence is a byproduct of how they already run: immutable audit logs, automated access-review reports generated on a schedule, and infrastructure-as-code that itself demonstrates the segmentation boundary, rather than a person manually assembling screenshots and spreadsheets every audit cycle.
Worked example
A company's two-year roadmap for a new payments feature might sequence this way: quarters 1-2, network segmentation isolating the new service into its own cardholder-data-environment subnet with encryption at rest and in transit designed in from the service's first deployment, not retrofitted; quarter 3, the formal access-review process and change-management gate go live before the service handles a single real transaction; quarter 4, a first internal PCI readiness assessment against the actual, running system; quarters 5-6, remediate whatever the readiness assessment found and complete the first formal PCI-DSS assessment (a Report on Compliance or a Self-Assessment Questionnaire, depending on the merchant's transaction volume, since which form applies is itself volume-dependent), with the evidence already produced automatically by the logging and access-review work built in quarter 3, not gathered by hand under deadline pressure. In parallel, GDPR data-subject request handling, a supported process to fulfill access and erasure requests within the required window, gets built alongside the same feature, since the same customer records are personal data under GDPR regardless of the payment angle that triggered the project.
Trade-offs and pitfalls
- Treating compliance as a checkbox exercise completed right before an audit produces brittle, one-time evidence instead of durable controls, and the same scramble repeats every audit cycle.
- Over-scoping the regulated boundary, putting more of the system "in scope" than actually needs to be, multiplies audit cost and slows down every unrelated change that happens to touch anything nearby.
- Compliance milestones that live on a separate tracker from feature milestones quietly get bumped every time a feature deadline is at risk, because nobody sees the collision until it's already happened.
Engineering leadership worries about vendor lock-in as the organization standardizes on one cloud provider. Propose a framework to quantify lock-in risk in business-relevant terms (not just technical ones) - covering egress fees, proprietary managed-service dependence, and estimated cost/effort to migrate later - and the architectural mitigations (abstraction layers, data-format standards, periodic migration drills) you'd recommend.
Sample Answer
Direct answer
Convert lock-in from a vague worry into a weighted numeric score built from a handful of business-relevant dimensions, then convert the score into an actual migration-cost-and-effort estimate in dollars and months. A number a finance stakeholder can compare against the value of staying is far more useful than a qualitative "high" or "low" risk label, and the scoring dimensions and mitigation catalog both need to flex by technology category, because a serverless function and an analytics platform fail differently.
Structured elaboration
Weighted scoring dimensions
Score each candidate service 1 (low risk) to 5 (high risk) on each dimension, multiply by its weight, and sum:
LockInScore=20×∑i=16wi⋅si
| Dimension | Weight | What it captures |
|---|---|---|
| Data egress cost exposure | 0.25 | Egress is the data that leaves the vendor's network; this dimension is the cost to move your data volume out at current pricing |
| Proprietary managed-service dependence | 0.25 | How much business logic lives inside vendor-specific managed features, not portable code |
| API/interface uniqueness | 0.15 | Whether an open standard or common interface exists as an alternative |
| Contractual constraints | 0.15 | Commitment length, termination penalties, minimum spend |
| Skills/tooling specialization | 0.10 | How much team expertise is tied to vendor-specific tooling |
| Ecosystem interdependency | 0.10 | How many other services from the same vendor this one is wired into |
The weights sum to 1, so the weighted average itself never leaves the 1-5 scoring range; the 20x multiplier rescales that 1-5 average into a 20-100 business-facing score, 20 if every dimension scores the safest possible 1, 100 if every dimension scores the riskiest possible 5, not a 0-100 score, since a weighted average of numbers that are never below 1 can never reach 0.
Technology-specific indicator sets
- Serverless/FaaS (function-as-a-service, a model where you deploy individual functions rather than servers): a proprietary event-trigger schema, an orchestration layer (a step-function-style workflow engine) that embeds business logic in vendor-specific configuration, and a per-invocation billing model that is hard to replicate elsewhere for cost comparison.
- Cloud analytics platform: a query dialect with non-standard SQL extensions, a storage format tied to the vendor's own compute engine, and data gravity (the sheer volume and the time and egress cost required to move it) as its own lock-in force independent of the API surface.
Cost and effort to migrate
MigrationCost=M⋅Ceng+E+D
where M is estimated engineer-months to rebuild the integration, Ceng is fully loaded monthly cost per engineer, E is one-time data egress cost, and D is the cost of running both systems in parallel during a dual-run validation window.
PaaS-specific scoring rubric
For a platform-as-a-service (PaaS) dependency such as a proprietary workflow-orchestration platform, score these four dimensions the same 1-5 way: API compatibility (can an equivalent platform run the same interface), data exportability (can workflow definitions and history be exported intact), custom runtime dependence (does your code rely on a runtime only this platform provides), and vendor-specific tooling dependence (deployment, monitoring, or debugging tools with no equivalent elsewhere).
Mitigations as a 24-month roadmap
| Phase | Action | Checkpoint |
|---|---|---|
| Months 0-6 | Introduce an abstraction layer around the highest-risk dependency; land canonical data in an open format | Abstraction layer covers the top-scoring service |
| Months 6-12 | Stand up a multi-cloud orchestration layer (for example, infrastructure-as-code and a container runtime that both major clouds support) for new workloads | New workloads default to the portable pattern |
| Months 12-24 | Run a live migration drill on the highest-risk existing dependency; refactor it if the drill reveals gaps | Drill completes within the estimated M months from the cost model |
Worked example
Scoring a proprietary FaaS platform: egress exposure 4, managed-service dependence 5, API uniqueness 4, contractual constraints 2, skills specialization 3, ecosystem interdependency 4. Weighted sum: 0.25(4)+0.25(5)+0.15(4)+0.15(2)+0.10(3)+0.10(4)=1.0+1.25+0.6+0.3+0.3+0.4=3.85, times 20 gives a LockInScore of 77 out of 100, solidly high risk. Migration estimate: rebuilding the function layer on a portable container runtime is estimated at 5 engineer-months at a fully loaded $18,000 per engineer-month ($90,000), plus $12,000 in one-time data egress, plus a 6-week dual-run window costing roughly $15,000 in duplicated infrastructure, for a total estimated migration cost of $117,000. That number, not the abstract "77," is what goes in front of a budget owner.
Trade-offs and pitfalls
The most common failure is scoring the dimension that is easiest to measure (egress cost, which is a line item) while under-scoring the one that matters more (managed-service dependence, which requires actually reading how deeply the code is coupled). A second failure is treating the score as permanent: usage and dollar exposure grow with adoption, so a service scored low at pilot scale should be re-scored as it becomes load-bearing. Finally, do not spend more on the scoring exercise than the position justifies; a low-weighted, low-volume service does not need the full rubric.
Design a vendor exit strategy for a mission-critical managed service you depend on. Cover the contractual, technical, and operational steps: data-export formats, API-compatibility fallback, contract freeze periods, dual-run testing, and the readiness to bring the service in-house if the vendor sunsets it or the relationship breaks down.
Sample Answer
Direct answer
A vendor exit strategy is only real if it has been rehearsed; a document describing what you would do if the vendor sunsets the service is not the same thing as knowing the export actually works. Cover it in three layers, contractual, technical, and operational, and underneath the technical layer, be specific about how a real data migration would actually execute under time pressure.
Structured elaboration
Contractual
- Explicit data-export rights: which format, at what frequency, at no additional cost.
- A reasonable termination clause with adequate notice, and a defined contract freeze period: a transition window during which the vendor guarantees no breaking API or pricing changes.
- Source or configuration escrow if meaningful business logic lives inside vendor-hosted workflow configuration, not just data.
Technical
- An API-compatibility fallback layer: an internal adapter your application already calls, so a replacement backend can be pointed to without touching every call site.
- Canonical data kept in an open, standard format alongside whatever the vendor's own optimized copy is.
- Dual-run testing: running the incumbent and a candidate replacement side by side and diffing outputs before ever committing to a cutover.
Operational
- A named owning team and a written runbook for bringing the capability in-house or switching vendors.
- Explicit trigger conditions that start the exit clock: a sunset notice, a price increase beyond an agreed threshold, or a defined number of SLA (service-level agreement) breaches in a rolling window.
The migration mechanics underneath the plan
When the trigger fires, the actual database migration has five concrete steps:
- Extraction: take an export at a defined checkpoint, choosing between a logical dump (slower, portable, verifiable row by row) and a block-level snapshot (faster, but tied to the source engine's format).
- Transformation: map schema and types explicitly, and flag every vendor-specific extension (a proprietary column type, a non-standard index) that has no direct equivalent, rather than silently dropping it.
- Online versus offline transfer: offline means pausing writes and bulk-copying, which is simple and fast but causes downtime; online means streaming ongoing changes via change-data-capture (CDC, a technique for capturing row-level changes as they happen) into the new system, followed by a short final cutover window, which is more complex but avoids extended downtime.
- Encryption and key handling: re-encrypt with your own key management system immediately on arrival; never let data sit unencrypted, even briefly, at the destination, and if the source vendor held the encryption keys, plan the key release or rotation as an explicit extraction step, not an afterthought.
- Integrity validation: row counts, checksums on a sorted export compared against the source, and a reconciliation query, all run before the old system is decommissioned, not after.
Worked example
A company depends on a proprietary managed workflow-orchestration vendor that issues a 12-month end-of-life notice. At month 12-before-deadline (T-minus-12), the notice is the trigger that starts the runbook. By T-minus-9, a dual-run environment is stood up against the chosen replacement. By T-minus-6, the first full dual-run diff test runs on a non-critical workflow, surfacing that the vendor's proprietary retry-backoff configuration has no direct equivalent and must be reimplemented explicitly, which is exactly the kind of gap the transformation step is meant to catch early rather than at cutover. By T-minus-3, the highest-volume workflow completes its own dual-run validation. At T-0, traffic cuts over behind a canary, with the old system kept running in a read-only, rollback-ready state for a defined 30-day bake period before decommissioning.
Trade-offs and pitfalls
Running a dual-run environment is real, ongoing, budgeted cost, not a free insurance policy; treating exit-readiness as costless is how it never actually gets built. The more common failure, though, is writing the contractual export right and never having engineering validate that the export is complete and actually loadable, so the plan looks solid on paper and fails the first time it is exercised for real. An untested runbook fails exactly when you need it most.
Your company must satisfy data-residency and encryption requirements that differ across the countries you operate in. As the technical decision-maker, how do you factor that multi-jurisdiction compliance requirement into your infrastructure choices (regions, key management, backup location) without fragmenting your architecture unnecessarily?
Sample Answer
Direct answer
Keep the application architecture shared wherever the law actually allows it, and localize only the specific data stores and encryption keys that a jurisdiction genuinely constrains; standing up an entirely separate regional stack per country by default is the expensive reflex most teams reach for first, and it is usually broader than what the law actually requires.
Structured elaboration
Classify by legal constraint, not by convenience
For each data category (personally identifiable information, payment data, health data, and so on), determine per country whether it faces a true residency requirement (data must physically stay in that region), a processing or transfer-adequacy constraint (data can leave, but access and processing must meet certain controls), or no special constraint at all. These are legally distinct, and conflating them is the most common cause of unnecessary fragmentation.
Architectural response by layer
- Regions: deploy a dedicated regional presence only where a genuine residency requirement exists; everything else stays on the existing shared regions.
- Key management: use a key management system supporting regional key isolation for the residency-constrained data specifically, so its encryption keys never leave that jurisdiction, while non-constrained data continues on the existing shared key hierarchy; this is usually far cheaper than replicating an entire database.
- Backup location: backups of residency-constrained data must also stay in-region; a commonly missed gap is solving primary-region residency and then routing backups to a shared, out-of-region disaster-recovery vault, which is often still a violation of the same requirement.
- Application tier: keep a shared control plane (the layer that manages configuration, deployment, and orchestration, not the customer data itself) and application logic where the law allows, and isolate only the data plane (where the actual customer data is stored and processed) where required, using a routing layer that directs a given tenant's data-plane calls to the correct jurisdiction's data store, so most of the system, application code, the deployment pipeline, and observability, stays unfragmented.
Worked example
A company serves customers in the European Union, which falls under the General Data Protection Regulation (GDPR); GDPR itself does not strictly mandate in-EU storage, it mandates adequate protection for any transfer outside the EU, an important distinction from a hard residency law. The same company also serves a customer in a country with an explicit data-localization law requiring certain data categories to stay in-country, a stricter model than GDPR's transfer-adequacy approach.
| Jurisdiction | Actual legal requirement | Architecture response |
|---|---|---|
| European Union (GDPR) | Transfer-adequacy, not strict residency | Keep data in the nearest compliant region and rely on an adequacy mechanism (such as standard contractual clauses) rather than forcing a dedicated EU-only deployment |
| Country with strict data-localization law | True in-country residency for the constrained data categories | Stand up a dedicated regional data store, regional key management, and a regional backup vault only for that jurisdiction's constrained categories; the application tier remains shared |
Trade-offs and pitfalls
Over-fragmenting by default, treating every country with any privacy law as if it required a full dedicated regional stack, multiplies the total cost of ownership and the ongoing operational burden for no actual legal benefit. Getting the legal distinction wrong yourself, rather than engaging counsel per jurisdiction, is a serious risk here specifically because whether a law requires strict residency versus transfer-adequacy is a legal determination, not an engineering judgment call; treat your own read of the requirement as a working hypothesis for legal review, never as the final answer. And the backups-in-the-wrong-region gap described above is, in practice, the most common real finding when this area gets audited.
Unlock Full Question Bank
Get access to all 8 Infrastructure Strategy and Technology Selection interview questions and detailed answers.
Sign in to ContinueJoin thousands of developers preparing for their dream job.