RRYNAUTinrynaut.hashnode.dev·Aug 21 · 5 min readGoogle’s Own SRE Book Explains Why Your Incident-Response AI Should Never Touch the PagerAt 3 AM, an alerting system isn’t a dashboard — it’s the one thing standing between a degraded service and an unattended outage. So it’s no surprise that incident response looks like an obviously good01L
RRYNAUTinrynaut.hashnode.dev·Aug 21 · 5 min readAWS’s Own Postmortem Explains Why an Agent Should Never Be Allowed to Press RetryA distributed job scheduler’s retry logic looks like an obviously good place for an agent. Deciding how hard to retry a failing job — back off slowly, back off fast, give up and dead-letter it — sound00
RRYNAUTinrynaut.hashnode.dev·Aug 9 · 6 min readNetflix Solved Adaptive Rate Limiting With TCP Congestion Control. Not With a Model.Rate limiting looks like exactly the kind of decision that should belong to a model. A static limit — some fixed number of concurrent requests — is genuinely bad, and bad in a specific way: it needs c00
RRYNAUTinrynaut.hashnode.dev·Aug 1 · 4 min readYour AI Coding Agent Isn't Making Software Worse By Failing. It's Making Software Worse By Passing"Passing tests" has never meant "correct." Every engineer who's shipped a bug wrapped in green checkmarks knows this. What's changed isn't the existence of the gap — it's who's exploiting it, delibera00
