Governance
85 bites tagged Governance — interview questions with model answers, and 60-second explainers.
Enforce a cloud resource compliance policy
Prevent at creation with org policies or admission checks, detect violations via continuous config scanning, and auto-remediate by stripping the IP or alerting owners. policy-as-code guardrails. manual periodic audits only.
Automating a no-public-IP governance rule
Use organization-level policy guardrails (SCP, Azure Policy, Org Policy) to deny public IP attachment before creation, applied across all accounts. preventive policy-as-code governance.
Measuring design system adoption and health
Component coverage and version spread from code scans, usage analytics, issue and satisfaction signals. quantifying adoption with concrete, automatable metrics.
Design systems across micro-frontends
Shared singletons and version alignment, token-based theming, distribution via shared scopes, stronger governance. adapting a design system to independently deployed front ends.
Governance for accessibility regressions
Detection via audits and monitoring, triage with severity and ownership, remediation SLAs, prevention feedback. a repeatable a11y governance loop across consuming products.
Explain event schemas and why schema registries matter at scale
This tests schema evolution and data contracts in distributed systems. A good answer defines schemas as contracts, explains that a registry enforces compatibility to block breaks, and lists pain like pipeline failures. Never treat schemas as optional docs.
Design a centralized participant management system to prevent over-contacting
Tests ResearchOps governance at scale. Strong answers define unified data tracking contact history, consent, and segments; enforce hard frequency caps and cooling-off windows; and build automated guardrails. Red flag: siloed spreadsheets or soft guidelines.
Research Governance Model
A research governance model is the operating system for scaling insights, trading autonomy for coordination so studies do not collide. It matters when teams share users and budgets.
How does a model registry differ from cloud storage like S3?
This tests model governance beyond raw storage. A strong answer contrasts storage with stage transitions, lineage, and ACLs, then lists metadata like metrics, dependencies, and schemas. A red flag is treating S3 folders with naming conventions as a registry.
Explain a model registry's purpose and what to store per version
Tests if you treat the registry as a governance bridge between training and production, not just storage. Strong answers cite versioned artifacts, lineage, metrics, dependencies, and approval gates. Red flag: calling it a file dump or experiment tracker.
Design a showback or chargeback system for ML infrastructure costs
Tag workloads to cost centers; define shared-resource formulas; automate reconciliation; use showback. Bridging ML telemetry with FinOps for shared GPU storage. Using raw cloud bills as attribution without GL mapping.
Why use a Model Registry over dated pickle files?
A strong answer covers versioning, lineage, promotion aliases, and governance. Why ad-hoc storage fails as models and teams scale. Treating deployment as a file copy rather than a managed lifecycle.
Model Risk Management: The Immune System for Production Models
Model Risk Management treats every deployed model as a liability that can silently decay. Banks use it to stop bad predictions from becoming bad decisions. The footgun is treating validation as a one-time checkbox instead of continuous governance.
Centralized vs Decentralized ML Platforms
A centralized ML platform trades team autonomy for standardization, while decentralized platforms embed ML tooling inside product teams. Centralized suites drown in ticket queues; decentralized ones duplicate cost and security holes without strong governance.
Model Registry: Source of Truth for Deployed Models
A model registry is the source of truth for which trained model runs where, turning anonymous artifact files into versioned, staged assets. It matters when you deploy multiple models or need instant rollbacks.
NIST AI RMF for LLM Deployment
The NIST AI RMF is a pre-flight checklist for organizational AI risk, not just code bugs. Teams use it to justify LLM deployment across legal, security, and fairness dimensions.
How do you handle a one-off token exception for a campaign?
Tests balancing speed and systemic consistency. Strong answers outline a governance fast-track with sunset clauses, document the exception, and isolate it via campaign overrides instead of mutating core tokens. Red flag: permanently hardcoding the one-off.
What governance model would you propose for a centralized design system?
Tests balancing standardization with autonomy across legacy silos. A strong answer proposes a hybrid model with a core team and contribution rules, weighing consistency against migration cost.
Hybrid Model: Core Team Plus Federated Contributors
A hybrid design system team pairs a central group owning primitives with product squads shipping patterns. Large orgs use it when one team cannot cover every surface. It fails when contribution rules are vague and the core becomes a bottleneck not an enabler.
Federated Design System Team Model
Federated ownership spreads design system control across teams, not a silo. A lean core owns standards while contributors ship components from the field. The footgun is treating federation as zero governance, splintering the system into forks.
Design System Roadmap: A Prioritization Contract
A design system roadmap is a contract with product teams, not a release calendar. It balances platform health against consumer demand so the system stays relevant. The footgun is letting roadmaps dictate every sprint, reducing the team to a service desk.
Centralized Design System Team Model
A centralized design system team acts as a central kitchen: one dedicated group owns the components and standards that every product team consumes. It guarantees consistency but becomes a bottleneck if the team loses touch with shipping product teams.
ML Model Registry: Source of Truth for Production Models
A model registry is version control for trained models, not just code. It tracks which artifact is running in production, who approved it, and how it was built. Skip it and you get untracked files in S3 with no way to reproduce a production model.
Design System Guild: Cross-Team Governance
A design system guild lets squads co-own the system, not a central team alone. It scales when one team cannot cover every surface. The footgun is voluntary membership without decision rights; it devolves into a committee that ships nothing.
Get Governance bites daily.
Five a day, five minutes, offline. With quizzes so it sticks.
Open testing — you’ll join as an early tester.