#architecture
9 posts tagged architecture. View all posts.
Is it worth buying a SaaS boilerplate in 2026?
The honest buy-vs-build answer for SaaS boilerplates — what you're actually paying for, the cases where a kit pays for itself, and the cases where you should walk away and build it yourself.
AGENTS.md and CLAUDE.md are table stakes — enforcement is the moat
Every serious SaaS boilerplate now ships an AGENTS.md or a CLAUDE.md — a soft file the agent can quietly ignore by prompt 50. The differentiator is enforcement: ESLint-checked DDD boundaries and typed contracts that turn the build red when the agent crosses a line.
MercadoPago webhooks done right: signature verification, idempotency, and the truth-lands-later problem
A walkthrough of UseDeploy's real MercadoPago adapter — HMAC signature verification with a timing-safe compare, verify-then-fetch webhooks, and idempotent read-model writes that survive at-least-once redelivery and the MercadoPago Checkout Pro sandbox.
Guardrails your agent can't write for itself: compiler-enforced DDD boundaries
Clean architecture survives contact with AI agents only when the boundaries are compiled in. Here are the exact ESLint rules UseDeploy uses to make DDD layering and cross-context walls something a coding agent can't cross by accident.
One billing port, three providers: Stripe, MercadoPago and Polar
How UseDeploy puts Stripe, MercadoPago and Polar behind a single typed IPaymentProvider interface — so charging in pesos or dollars is a one-file swap, and the rest of the app never sees a provider SDK.
Why AI coding agents keep breaking things that already worked
AI generated code technical debt has a shape — agents drift toward their own defaults and quietly break code that already shipped. Why it happens, and what actually holds the line.
Signed, retryable outgoing webhooks: HMAC, an outbox, and BullMQ
How UseDeploy delivers outgoing webhooks your customers can actually trust — Stripe-style HMAC signatures, a durable transactional outbox for the fanout, and BullMQ for per-delivery retries.
Multi-tenancy in Prisma without RLS: a tenant guard that fails loud
How UseDeploy stops cross-tenant data leaks with an org-scoped Prisma query extension that throws when a query forgets its organizationId — no Postgres RLS required.
Reusing one multi-tenant base across every client without rebuilding it
How a multi-tenant SaaS boilerplate turns the org/membership/role model into something you build once and ship to every client — system roles, per-org permissions, and the invariants a fresh rebuild always forgets.