Join The CTO Playbook Slack Community to connect with other CTOs!

What if the most complex part of your backend system... just vanished?

In this episode, Adam sits down with Maxim Fateev, CTO and co-founder of Temporal, a platform quietly revolutionizing how distributed systems are built at companies like Stripe, Snap, and Netflix. With a career spanning Amazon, Google, Microsoft, and Uber, Maxim brings a rare, battle-tested perspective to the chaos of modern infrastructure.

This conversation unpacks why the old way of building event-driven systems is fundamentally broken—and how Temporal’s concept of durable execution makes reliability and scale feel effortless. You’ll learn why your devs are wasting thousands of hours on boilerplate, how workflows are being redefined by AI, and why most “modern” architectures are really just legacy spaghetti in disguise. If you care about developer velocity, rock-solid uptime, or building tech that doesn’t implode at scale, this one’s a must-listen.

You’ll Learn:

How durable execution eliminates retries, race conditions, and orchestration chaos
Why event-driven systems create hidden coupling—and how to break it
How Amazon’s early monolith shaped a new paradigm for backend scale
What makes step functions a step backwards in modern architecture
How to write workflows that survive failures, restarts, and region-wide outages
Why most business logic doesn’t need YAML, queues, or 12 microservices
How to scale from prototype to production without rewriting core systems
What AI agents reveal about the future of workflow orchestration
How Temporal boosts developer productivity by 5–10x in real-world teams
Why durable execution is becoming the backbone of enterprise AI systems

Timestamps:

[00:00] Introduction

[00:37] Why durable execution matters in distributed systems

[03:36] The pain of monoliths and the shift to microservices

[05:19] Why events create tight coupling in architecture

[07:00] Amazon’s early attempt at workflow orchestration

[07:50] Why AWS step functions fall short

[09:08] How durable execution actually works

[10:44] Open-source vs cloud-hosted Temporal

[12:48] Multi-region replication and total uptime

[13:58] Write workflows in code, not configs

[15:50] Is Temporal just for big companies?

[16:37] 5–10x productivity gains with durable execution

[18:26] Durable execution as a new middleware layer

[20:03] Replacing bloated cloud architecture with one function

[21:10] When not to use Temporal

[23:03] Netflix’s switch from Spinnaker to Temporal

[24:15] Language support and runtime challenges

[25:03] New features: priority queues and fairness

[26:36] Nexus: long-running RPC calls with full durability

[28:34] Getting started with Temporal

[30:19] Why Max stepped down as CEO

[32:14] Durable execution powering AI workloads

[35:18] Why AI agents need orchestration

[37:03] Saving state between agent calls

[39:34] Durable execution vs traditional AI workflows

Resources Mentioned:

Amazon Simple Workflow Service | Website

AWS Step Functions | Website

Temporal (Open Source) | Website

Temporal Documentation and Courses | Website

Azure Durable Functions | Website

Temporal Slack Community | Website

If you want to learn more about Max’s work, follow him on LinkedIn.

Find more from Adam on LinkedIn and YouTube, and check out Adam's CTO coaching company Synova Tech.

Kumail Nanjiani on World Cups, Hollywood and India-Pakistan Relations

The Grade Cricketer

Conan, cricket exposure and celebrity interest

The AI-powered Podcast Player