Kubai Kevin

32 articles tagged Identity & Access

RBAC vs ABAC: pick by blast radius

RBAC vs ABAC for services: why role explosion and attribute spoofing fail in opposite directions, with runnable policy code and a deny-on-missing test.

12 min read · 17 November 2026

Saga pattern: orchestration vs choreography in payments

How payment sagas coordinate steps, why refunds are not charge reversals, and a decision framework for choosing orchestration or choreography.

16 min read · 10 November 2026

Envelope encryption with cloud KMS: key rotation traps

How envelope encryption with cloud KMS breaks during key rotation, why wrapped data keys and key IDs must travel with ciphertext, and how to test it.

12 min read · 6 November 2026

Gateway API won't fix your Ingress mess

Gateway API does not fix messy Ingress manifests. It changes who owns routing. Here is when delegation is worth the migration and when to wait.

13 min read · 2 November 2026

STRIDE is a checklist, not a threat model

STRIDE is a memory aid, not a threat model. Use attack trees to map paths to an asset and STRIDE to prompt each node, with a CI check that keeps trees current.

15 min read · 1 November 2026

JWT sessions: revocation, rotation, storage

Why JWT revocation is a state problem, how refresh token rotation with reuse detection works, and how token storage shapes the threat model.

14 min read · 27 October 2026

7 guardrails for safer LLM apps

A practical review of seven guardrail layers for LLM applications: what each catches, where it fails, and which to build first.

18 min read · 25 October 2026

OAuth 2.1 migration: what breaks in existing integrations

OAuth 2.1 removes the implicit and password grants, mandates PKCE, and requires exact redirect URIs. Here is what breaks and how to sequence the migration.

18 min read · 22 October 2026

Docs that survive: the bus factor fix

How to write runbooks and design docs that a new engineer can follow without asking the author: explicit audience, exact commands, expected output, and rollback.

15 min read · 21 October 2026

MCP vs A2A: pick the wrong layer

MCP connects models to tools; A2A connects agents to agents. How to tell which layer a boundary belongs to, and what breaks when you mix them.

16 min read · 20 October 2026

PCI DSS architecture: 12 requirements that matter

Map PCI DSS requirements to concrete architecture decisions: CDE boundary, envelope encryption and key separation, immutable logging, CI/CD controls, and scope reduction.

14 min read · 19 October 2026

Read replicas: the bug factory you built

Read replicas lag by design. Learn how to classify reads, route freshness-critical queries to the primary, and measure your exposure to stale reads.

15 min read · 18 October 2026

Self-service AI tooling: the permission trap

Design self-service LLM layers so identity, context and budget controls prevent quiet permission failures instead of only working on the happy path.

14 min read · 9 October 2026

Who signs off on AI code?

AI coding assistants produce diffs fast, but accountability still needs a named human. Here is a review workflow with hooks, policy checks, and failure modes.

12 min read · 2 October 2026

Settlement vs clearing: backend engineer's guide

Clearing produces obligations; settlement moves money. How to model both ledgers, reconcile them, and avoid the schema mistakes that break finance.

15 min read · 28 September 2026

Prompt injection: what breaks first in prod

Why prompt injection defeats prompt-level defenses, and how to bound the blast radius with tool allowlists, argument scoping, URL filtering and confirmation gates.

15 min read · 27 September 2026

Agent compliance: the layer you build after failure

Pre-execution policy gates miss most agent compliance failures. Build a decision trace that reconstructs what happened, why, and what would have stopped it.

16 min read · 22 September 2026

When your agent goes silent: degrade gracefully

How to keep agent-backed features useful when a model is slow or confidently wrong: deadlines, fallbacks, circuit breakers, validation, and how to measure each.

13 min read · 21 September 2026

Computer-use agents: the permission trap

GUI agents inherit a human's full permissions and leave no semantic audit trail. Here is how a policy layer, a reduced interface and structured logs fix that.

15 min read · 19 September 2026

Scope agent permissions per task

Most agent permission models fail because credentials are scoped to the agent, not the task. Here is how to issue short-lived, task-scoped tokens and what to measure.

20 min read · 19 September 2026

Zero-trust AI: skip the latency tax

How to keep per-request identity and least privilege in inference services while moving handshakes and policy evaluation off the hot path.

14 min read · 15 September 2026

Agent auth: where the docs ended up lying

OAuth2 and OIDC assume reliable clocks, stable power and trusted hardware. Here is what breaks on constrained agent networks, and how to measure it.

14 min read · 24 August 2026

Least privilege for agents: the config line nobody writes

Why permissive agent tool defaults break in production, and how to add runtime, context-aware policy checks before every tool call.

14 min read · 10 August 2026

Prompt Injection Is Not Your Biggest LLM Data Risk

Prompt injection dominates LLM security checklists, but poisoned training data and label drift often cause more damage. A pipeline-stage model for triage.

17 min read · 3 August 2026

Agent auth: the silent deployment killer

How agent identity breaks under jitter, power loss, and low-end devices, and the fallback, rotation, and caching patterns that keep auth working.

13 min read · 30 July 2026

M-Pesa Daraja 2.0: your 2026 migration checklist

A practical checklist for migrating an M-Pesa Daraja integration across a breaking API change: auth, idempotency, callbacks, error handling and rollout.

14 min read · 2 July 2026

Platform teams in 2026: Docs vs. production reality

Platform docs describe happy paths; production exposes the gaps. Covers cluster defaults, IDP design by company size, Helm guardrails, and policy enforcement.

18 min read · 27 June 2026

Who owns AI-generated code?

AI can write code, but a human still signs off on behavior. Here is how to make code ownership explicit with prompts, tags, and a security gate.

10 min read · 27 June 2026

Claude Code after a year: daily wins

Design an agentic Dependabot patcher that bumps a dependency, runs tests in a container, and opens a PR only when the suite passes.

13 min read · 25 June 2026

Istio and Cilium: what really breaks in production

Service meshes amplify latent platform failures. Here is what breaks in Istio and Cilium, how to measure it, and when to defer adoption.

13 min read · 24 June 2026

Replace .env with these in 2026

Why .env files fail in production and how managed secret stores, vault agents and encrypted-file workflows replace them, with rotation and caching tradeoffs.

15 min read · 18 June 2026

7 code review traps when AI writes half your code

Seven review traps AI-generated code creates, how to detect each one, and a checklist you can add to your repo today.

12 min read · 11 June 2026