Prismatic logo
Headshot of Brian Munz

Brian Munz

Developer Advocate

Brian brings over twenty years as a developer and more than a decade in developer relations to Prismatic. As founding developer advocate, he builds the videos, workshops, and demos that show developers how to extend their products with code-first integrations, embedded workflows, and tools that AI agents can act on.

Posts by this author

How Do I Implement Exponential Backoff with Jitter
Dev Tips
How Do I Implement Exponential Backoff with Jitter That Respects Providers' 429 Semantics?Learn why "retry with backoff" isn't enough for integrations, and how to build backoff with jitter that reads each provider's rate-limit signals correctly.
Mean Time to Integration Recovery
Metrics
Mean Time to Recovery: How Quickly Can You Restore an Integration?Learn how to measure mean time to recovery (MTTR) for customer-facing integrations and use incident data to improve diagnosis and restoration speed.
How Do I Refresh OAuth 2.0 Tokens at Scale
Dev Tips
How Do I Refresh OAuth 2.0 Tokens at Scale without Race Conditions Across Parallel Requests?Master OAuth 2.0 token refreshes at scale. Avoid concurrency bugs, handle token rotation safely, and implement distributed locks correctly in production.
Mean Time to Detect Integration Failures
Metrics
Mean Time to Detect: How Quickly Will You Know an Integration Failed?Learn how to measure mean time to detect (MTTD) for customer-facing integrations and separate alert delivery from team awareness.
How Do I Make Webhook Processing Idempotent
Dev Tips
How Do I Make Webhook Processing Idempotent So Retries Don't Create Duplicates?Discover why webhook providers send duplicate events and learn step-by-step how to make your integration processing safe, robust, and fully idempotent.
How to Measure the Work Your Customers Expect
Metrics
How to Measure the Work Your Customers ExpectLearn how to calculate the expected execution success rate for integrations. See why platform uptime can look good while hiding integration failures.
Who One-Shot Ya?
Who One-Shot Ya? Why the One-Shot Benchmark Isn’t UsefulThe one-shot benchmark is the wrong way to judge AI agents. Most failures aren't model failures; they're context failures. Here's what to measure instead.
What AI Can and Can't Take off Your Integration Team's Plate
Dev Tips
What AI Can and Can't Take off Your Integration Team's PlateWrite integration code faster with AI, but don't ignore multi-tenancy, auth, and ops. See how AI and Prismatic team up to scale your B2B SaaS integrations.
How Developers Use AI to Build Integrations Faster
Dev Tips
How Developers Use AI to Build Integrations FasterAI enables faster integration development. Prismatic helps it survive production with dev-friendly tools and an infrastructure that scales with you.

Stay in the loop

Subscribe to our newsletter to get this content straight to your inbox.
Subscribe