Cursor & Co.
Engineering notes

How we build integrations, platforms, and automation

Written by the engineers doing the work: architecture patterns, failure modes, and the cost drivers behind integration and platform projects. No vendor marketing, no hand-waving — the decisions we actually make on engagements.

Requests routed to the right destination
Delivery metrics trending upward
Featured · Integration engineering

Freshworks CRM integration: architecture patterns that hold up in production

Freshworks CRM integrates cleanly over its REST API plus webhooks. The durable pattern is: OAuth or API-key credentials held server-side, webhooks for near-real-time change capture, a scheduled reconciliation sweep as the safety net, field-level ownership rules to resolve conflicts, and an idempotent event log so every write can be replayed without duplication.

14 Jul 202611 min read
All articles

Engineering notes archive

Data moving through a staged pipeline
Integration engineering

Zoho integration patterns: CRM, Books, and Desk without the spreadsheet in the middle

How to integrate Zoho CRM, Zoho Books, and Zoho Desk with products, databases, and payment providers: API surfaces, OAuth, metered billing, and month-end close safety.

23 Jun 2026 · 10 min read

ZohoZoho BooksBilling
Dashboards watching throughput and errors
Integration engineering

ServiceNow integration patterns for engineering teams

Table API versus Scripted REST, incident bridging, CMDB alignment, SLA-safe state mapping, and loop suppression when connecting ServiceNow to engineering tooling.

2 Jun 2026 · 10 min read

ServiceNowITSMIncident management
Records migrated from a legacy system to a new one
Integration engineering

How to build webhook consumers that survive retries, duplicates, and outages

A production checklist for webhook receivers: signature verification, idempotency keys, ordering, replay, dead-letter handling, and the observability that makes failures visible.

19 May 2026 · 9 min read

WebhooksIdempotencyReliability
An API surface documented endpoint by endpoint
Integration engineering

Bidirectional sync: conflict resolution strategies that hold up in audit

Field-level ownership, vector clocks, CRDT-style convergence, and reconciliation sweeps — how to keep two systems agreeing without last-write-wins data loss.

5 May 2026 · 10 min read

SyncData integrityArchitecture
Data moving through a staged pipeline
Platform engineering

CRM to data warehouse: getting numbers your board will trust

Extraction patterns, slowly changing dimensions, metric ownership, freshness monitoring, and lineage when moving CRM data into Snowflake, BigQuery, or Postgres.

21 Apr 2026 · 9 min read

Data platformSnowflakeReporting
Dashboards watching throughput and errors
Engagement & cost

iPaaS or custom integration? A decision framework, not a preference

When Zapier, Make, Workato, or a vendor connector is the right answer, when custom engineering pays for itself, and the total-cost factors most teams miss.

7 Apr 2026 · 9 min read

iPaaSBuild vs buyCost
Records migrated from a legacy system to a new one
Engagement & cost

What a custom software integration actually costs — and why estimates vary so much

The cost drivers behind integration projects: object count, sync direction, data quality, auth model, volume, compliance, and the discovery work that prevents overruns.

24 Mar 2026 · 8 min read

CostScopingDelivery
An API surface documented endpoint by endpoint
Platform engineering

Designing for API rate limits: budgets, back-off, and backfills that do not starve production

Token buckets, shared request budgets, exponential back-off with jitter, priority lanes, and why naive retries make outages worse.

10 Mar 2026 · 8 min read

ReliabilityRate limitsAPIs
Data moving through a staged pipeline
Platform engineering

Multi-tenant SaaS architecture: the decisions that are expensive to reverse

Tenant isolation models, entitlement as data, metering, per-tenant limits, and the tests that prove a query cannot cross a tenant boundary.

24 Feb 2026 · 10 min read

SaaSMulti-tenancySecurity
Dashboards watching throughput and errors
AI engineering

Reviewable AI automation: shipping LLM features operations teams actually keep

Confidence gating, labelled evaluation sets, regression harnesses, decision logs, and reversibility — the engineering that separates durable AI automation from an abandoned pilot.

10 Feb 2026 · 9 min read

AIEvaluationAutomation
Records migrated from a legacy system to a new one
Platform engineering

Legacy modernisation without a big-bang rewrite

Applying the strangler-fig pattern to legacy business systems: seams, anti-corruption layers, dual-write and dual-read phases, data migration, and safe cutover.

27 Jan 2026 · 10 min read

ModernisationArchitectureMigration