Skip to main content

Platform Operations

What does it take to keep a platform running — and how do you make it better every cycle?

Platform operations is the discipline that sits between engineering and infrastructure. Engineering builds features. Operations ensures they deploy, run, and recover. Without operations discipline, every feature is a liability.

Operations Map

Three concerns, each with distinct ownership:

ConcernQuestionWhere It Lives
BuildHow does code become a deployed artifact?CI Testing Infrastructure
RunWhere does it execute and how is it monitored?Cloud Orchestration
VerifyHow do we know it works and what it costs?Testing Economics

Decision Boundaries

Every platform has positions to fill. These are the active decisions:

DecisionQuestionCurrent Choice
Cloud ServicesWhere does it run?Vercel (Web2) + Fleek (Web3)
Identity StackHow do you prove who someone is?Clerk + zkLogin
Data ORMHow do you talk to the database?Drizzle
Message BusHow do services communicate?Kafka + Azure Service Bus
File ManagementHow do you manage files?Evaluating
Payment RailsHow do you collect money?Stripe + Crypto

See Tech Stack for the full positions table and Tech Decisions for the evaluation process.

Dig Deeper

Context

Questions

What operations task are you doing manually today that should be automated infrastructure tomorrow?

  • Which deployment failure mode has cost you the most time in the last month?
  • If your CI pipeline disappeared, which quality gate would break first?
  • What's the gap between your test environment and production — and what bugs live in that gap?