The complete collection of Deep Dives, practical tutorials, Quick Reads and engineering notes across Salesforce, MuleSoft, DataWeave, integration, APIs, distributed systems, testing and developer tooling.
Looking specifically for longer-form architecture and production engineering? Explore the 15 curated Deep Dives →
Design SOQL for Salesforce large data volumes using selectivity, indexes, Query Plan analysis, stable pagination, Bulk API extraction patterns, and architecture-aware tradeoffs.
Read article →An architecture deep dive into production-grade MuleSoft–Salesforce integrations: API selection, synchronous and bulk patterns, CDC/Pub/Sub, idempotency, replay, resilience, security, observability, and recovery.
Read article →An architecture deep dive into reducing CDC noise, carrying correlation context, consuming with Pub/Sub API, and designing replay, idempotency, backpressure, and recovery.
Read article →A production-oriented error-handling strategy for Salesforce integrations in MuleSoft, separating transient connectivity failures from validation, authorization, and business errors.
Read article →A practical observability model for MuleSoft Salesforce integrations covering correlation IDs, latency, Salesforce errors, event lag, replay checkpoints, bulk jobs, and actionable alerts.
Read article →How to use Salesforce Connector upsert and external IDs to build retry-safe MuleSoft integrations without creating duplicate Salesforce records.
Read article →A practical review of Salesforce CDC field enrichment, including the 10-field limit, payload semantics, and when a follow-up query is still appropriate.
Read article →Compare carrying stable context in Salesforce CDC enriched fields with issuing a Salesforce query after each change event.
Read article →Combine Salesforce CDC filtering and field enrichment to build a smaller, more self-contained event stream for downstream integrations.
Read article →A balanced architecture review of Salesforce event filtered channels: where they help, where they add coupling, and what they do not solve.
Read article →Compare Salesforce server-side custom-channel filtering with receiving a broad event stream and filtering it inside the integration client.
Read article →Connect Salesforce CDC custom channels to a production Pub/Sub API subscriber with correct Avro decoding, replay, flow control, and recovery thinking.
Read article →Practical SOQL performance guidance for selective filters, indexed fields, leading wildcards, relationship queries, and incremental integrations.
Read article →A practical architecture and implementation guide for consuming Salesforce Change Data Capture events with MuleSoft while designing replay, idempotency, recovery, and downstream processing deliberately.
Read article →A practical architecture pattern for isolating failed integration work, preserving appropriate recovery context, and enabling safe replay without turning transient failures into permanent data loss.
Read article →A practical architecture guide to API governance across design standards, ownership, contracts, security policies, traffic controls, observability and lifecycle management.
Read article →Experience, Process and System APIs are architectural responsibilities, not a requirement to add three network hops to every integration.
Read article →A production architecture guide to protecting event consumers and downstream systems when event arrival rates exceed processing capacity.
Read article →Event producers and consumers evolve independently. This deep dive explores additive change, semantic compatibility, versioning and migration without turning the event bus into a coordination bottleneck.
Read article →A production design for consuming high-volume Salesforce Change Data Capture events without depending on the 72-hour event-bus window: durable handoff, replay checkpoints, sequencing, parallel processing, backpressure and recoverable error handling.
Read article →A practical architecture guide to preventing duplicate business effects when APIs, events, batches, and recovery workflows are retried or replayed.
Read article →A practical architecture for tracing business transactions across APIs, events, queues and downstream systems without confusing logs with observability.
Read article →A practical architecture guide to reliable event publication and consumption when business state, brokers, databases, and external systems cannot share one atomic transaction.
Read article →A practical architecture for consuming Salesforce CDC and Platform Events reliably, with replay checkpoints, idempotency, recovery paths, observability, and downstream failure handling.
Read article →Retries are only one recovery mechanism. This deep dive separates retry, replay and reprocessing and shows how to design each without creating duplicate business effects.
Read article →A practical guide to choosing MuleSoft processing patterns for large datasets, including batch jobs, aggregation, concurrency, memory, API limits, failure recovery, and observability.
Read article →A practical introduction to MuleSoft Object Store for watermarks, replay state, tokens, and lightweight application state - and when a database is the better design.
Read article →A practical guide to choosing Parallel For Each or Batch Processing based on payload size, synchronous completion, memory, concurrency, and record-level recovery needs.
Read article →Rate limits protect consumption over a time window; burst protection addresses sudden traffic. Understand the architectural distinction before choosing an API policy.
Read article →A practical framework for choosing between retry, replay, dead-letter queues, and an Error Hospital based on failure semantics, recovery needs, and operational complexity.
Read article →Salesforce Pub/Sub API is pull-based. Use its event-request model as a real backpressure mechanism instead of building an ever-growing consumer backlog.
Read article →