
TanStack Start or Next.js: Follow One Mutation Through Production
Both frameworks can render pages and execute server-side work. A more useful comparison follows one authenticated mutation through cache invalidation, browser state, deployment skew, and the actual hosting path.
Read full article →
Freelance Architect or Agency: Who Keeps the Decisions?
Choosing external engineering support is partly about capacity, but the harder questions concern decision ownership, continuity, review authority, and what remains usable after the engagement ends.
Read full article →
What a Skill Changed, and What It Didn't
A successful task can depend on a skill, existing instructions, tool access, tests, or a person correcting the assumption. Evaluating agent workflows means keeping those contributions apart instead of turning one good result into a productivity claim.
Read full article →
The Backup Exists. The Restore Still Needs Work.
A restored database can disagree with payment providers, object storage, jobs, and deployed code. Recovery is a coordinated system change, with controlled side effects and evidence that the recovered state is usable.
Read full article →
A Review Belongs to a Commit
Review automation has to track more than whether it has seen a PR. Findings, CI evidence, duplicate suppression and guarded actions need a revision and a lifecycle, while final approval remains an accountable human decision.
Read full article →
The Dashboard Is Green. The Order Is Missing.
Useful observability connects browser intent, API acceptance, durable state, and background outcomes. Sampling, cardinality, missing denominators, privacy, and telemetry failure determine which questions that connection can answer.
Read full article →
The Command Succeeded. The Scope Changed.
Hooks and generators can transform a small, reviewed change into a different committed artifact. Agent workflows need to inspect the output of automation, distinguish unrelated drift, and stop when the final scope no longer matches the task.
Read full article →
More Replicas, the Same Database
Horizontal scaling multiplies connection pools, retries, and deployment overlap as well as capacity. The useful questions concern admission limits, health-check meaning, draining, and which state a replacement instance can actually recover.
Read full article →
Green Checks Only Cover What You Checked
A local build can accept a dependency change while a related service rejects its contract. The useful work is establishing the scope and freshness of the evidence, including when the original proposal should be closed rather than completed.
Read full article →
A Small AWS Deployment, With the Awkward Parts Included
CloudFront, ALB, ECS, RDS, and an optional Redis layer are enough to explain a useful deployment. The less obvious work is controlling identity, cache behavior, outbound access, connection budgets, and the cost of keeping the system recoverable.
Read full article →