# Decision Log

_2 decisions extracted._

## Weighing two candidates for the senior backend opening

- **Date:** 2026-01-14
- **Chose:** Alex
- **Confidence:** high
- **Source:** claude · _Hiring — Sam vs Alex for the senior backend role_

> assistant: For a team of 8 where you already have distributed-systems depth in other hires, the marginal utility of Alex's shipping habit and communication is probably higher. Systems skill compounds; communication unblocks.
> user: Yeah that's the argument I'd been circling. Decided on Alex. Writing up the offer now.

---

## kubernetes vs fly.io

- **Date:** 2026-02-03
- **Chose:** kubernetes
- **Rejected:** fly.io
- **Tags:** `infra`
- **Confidence:** medium
- **Source:** claude · _Kubernetes or Fly.io for the next deploy_

> assistant: If the rest of infra is on k8s and CI is wired, the marginal cost of adding one more service is tiny — the cognitive and ops burden is amortised. Fly's nice-to-haves (scale-to-zero, simple UX) don't offset two new things to learn.
> user: Agreed. Sticking with kubernetes. The scale-to-zero thing was cool but it's a tiebreaker not a deciding factor.

---