HackerTrans
TopNewTrendsCommentsPastAskShowJobs

selenehyun

no profile record

Submissions

Separating Business Definition from Execution in Infrastructure

lynq.sh
1 points·by selenehyun·4 mesi fa·0 comments

In the Age of AI Agents, How Should Infrastructure Change?

lynq.sh
1 points·by selenehyun·5 mesi fa·0 comments

Show HN: A dashboard for understanding Kubernetes operators

lynq.sh
2 points·by selenehyun·7 mesi fa·0 comments

[untitled]

1 points·by selenehyun·7 mesi fa·0 comments

[untitled]

1 points·by selenehyun·7 mesi fa·0 comments

Lynq just added KillerCoda demos you can try in the browser

github.com
1 points·by selenehyun·7 mesi fa·1 comments

Show HN: Lynq – Turn Database Rows into Kubernetes Infrastructure Automatically

lynq.sh
1 points·by selenehyun·8 mesi fa·0 comments

Show HN: Tenant Operator – Provision Kubernetes Tenants Directly from Your DB

docs.kubernetes-tenants.org
4 points·by selenehyun·8 mesi fa·1 comments

comments

selenehyun
·7 mesi fa·discuss
I'm the solo developer behind Lynq. I originally wanted to offer a live demo that would provision a short-lived tiny app in real time so people could actually watch it boot and vanish. It was a fun idea, but running that safely on small homelab-style infrastructure didn’t feel solid enough for security or traffic spikes.

So I switched to providing the demo through KillerCoda instead. It gives a clean, isolated environment where you can try the workflow without setup. Just sharing in case it's useful.
selenehyun
·8 mesi fa·discuss
It is not open source, but I am still proud of a message delivery system I designed and built alone two years ago. It consists of six independent components and guarantees at least one successful delivery as long as the database remains available. It supports AWS SES, Twilio SMS and MMS, Webhook, Discord messages, and can easily add new providers through an adapter pattern.

Messages are queued through an API, captured by Debezium, produced to Kafka, delivered by workers, logged, and updated through DSNs received via webhook. Failures go to a DLQ where they are retried until the limit is reached.

Each stage runs independently, so any failure only causes minor delay without risking unintended drops. With Prometheus metrics in place, this system has processed more than two hundred thousand messages per day in production for two years without a single reported loss.
selenehyun
·8 mesi fa·discuss
Ref. https://news.ycombinator.com/item?id=45911851
selenehyun
·8 mesi fa·discuss
[dead]