Hacker News: chw9e

New comment by chw9e in "Subscription bombing and how to mitigate it"

chw9e — Thu, 02 Apr 2026 06:50:55 +0000

This happened to me several years ago. I got signed up to probably 700 newsletters overnight. In the middle of all of the sign ups there was activity on my airbnb account where my notification settings were changed. when i checked my airbnb i noticed that someone had created a fake listing under my account and disabled booking notifications for it. a real multi-layer scam where the hacker would be making money off a fake listing on someone else's account who would probably never even realize it.

Background Coding Agents: Predictable Results Through Strong Feedback Loops

chw9e — Tue, 03 Mar 2026 23:20:13 +0000

Article URL: https://engineering.atspotify.com/2025/12/feedback-loops-background-coding-agents-part-3

Comments URL: https://news.ycombinator.com/item?id=47240529

Points: 2

# Comments: 0

Agent-written tests can't verify agent-written code

chw9e — Mon, 02 Mar 2026 18:21:29 +0000

Article URL: https://qckfx.com/blog/agent-written-tests-cant-verify-agent-written-code

Comments URL: https://news.ycombinator.com/item?id=47221866

Points: 3

# Comments: 0

Diffusion Models for ARC-AGI: A Retrospective

chw9e — Thu, 12 Feb 2026 23:23:21 +0000

Article URL: https://christopherhwood.com/Diffusion-Models-for-ARC-AGI-A-Retrospective/

Comments URL: https://news.ycombinator.com/item?id=46996724

Points: 1

# Comments: 0

AI Agents are shipping code faster than we can test

chw9e — Mon, 09 Feb 2026 17:38:31 +0000

Article URL: https://christopherhwood.com/AI-Agents-Shipping-Code-Faster-Than-We-Can-Test/

Comments URL: https://news.ycombinator.com/item?id=46948204

Points: 4

# Comments: 0

Building iOS UI with Coding Agents Is Slow

chw9e — Tue, 13 Jan 2026 16:09:18 +0000

Article URL: https://qckfx.com/blog/building-ios-ui-with-coding-agents-is-slow-heres-how-to-fix-it

Comments URL: https://news.ycombinator.com/item?id=46602901

Points: 1

# Comments: 0

Show HN: qckfx – Stop manually checking if your AI agent broke your iOS app

chw9e — Thu, 08 Jan 2026 14:59:03 +0000

AI coding agents can build iOS UI, but they can't verify it. They can update a screen, but the UI might drift or break, and nobody catches it until a human checks.

qckfx gives your agent a baseline. Record a simulator session once. Every tap, scroll, and network response gets captured. On replay, each screen is compared against the original.

With our MCP, your agent triggers the tests and gets back visual diffs of exactly what changed. Updating the baselines is one click.

Under the hood:

- Full network replay (HTTP & WebSocket)

- Initial disk & keychain state captured during recording and restored on every run

- Precise scroll positioning (built from scratch; XCUITest only exposes this on macOS and iPad)

- No AI in the loop at runtime, fast execution

No SDK or code changes needed. Nothing to commit to git. Just download the app and go.

Everything runs locally. Your data stays on your machine.

https://qckfx.com

Comments URL: https://news.ycombinator.com/item?id=46541792

Points: 2

# Comments: 0

A Guide to Hiring the Illegible

chw9e — Tue, 06 Jan 2026 02:16:33 +0000

Article URL: https://www.optimaloutliers.com/p/a-guide-to-hiring-the-illegible

Comments URL: https://news.ycombinator.com/item?id=46507938

Points: 2

# Comments: 0

Giving Your Agent Eyes Is Not Enough

chw9e — Fri, 02 Jan 2026 18:53:09 +0000

Article URL: https://qckfx.com/blog/giving-your-agent-eyes-is-not-enough

Comments URL: https://news.ycombinator.com/item?id=46468044

Points: 1

# Comments: 0

Show HN: qckfx – Record your iOS simulator sessions, replay them as tests

chw9e — Tue, 23 Dec 2025 23:27:36 +0000

I built this because I hate writing tests but I also hate shipping bugs.

qckfx records what you're already doing. You tap through the simulator to check your work, and it replays those sessions to catch visual regressions and crashes. No SDK, no code changes, no test files.

How it works: finish a feature, click around in the simulator, hit Cmd+Shift+S. That's your test. Cmd+Shift+T to run it before you push.

It's free, runs locally, and installs via Homebrew: brew install --cask qckfx/tap/qckfx. I'm a solo dev (10 years iOS, ex-Meta) and this is early. Feedback welcome.

https://qckfx.com

Comments URL: https://news.ycombinator.com/item?id=46370721

Points: 2

# Comments: 0

We Use API Agents to Build Integrations Fast

chw9e — Fri, 05 Dec 2025 17:18:06 +0000

Article URL: https://qckfx.com/blog/how-we-use-api-agents-to-build-integrations-fast

Comments URL: https://news.ycombinator.com/item?id=46164249

Points: 1

# Comments: 0

What I Wish I Could Tell Every Engineer Joining a New Team

chw9e — Thu, 04 Dec 2025 16:09:37 +0000

Article URL: https://qckfx.com/blog/what-i-wish-i-could-tell-every-engineer-joining-a-new-team

Comments URL: https://news.ycombinator.com/item?id=46149154

Points: 1

# Comments: 0

Dropbox: Knowledge Graphs, Prompt Optimizers, and MCPs

chw9e — Thu, 04 Dec 2025 01:53:08 +0000

Article URL: https://maven.com/p/7d35a7/dropbox-knowledge-graphs-prompt-optimizers-and-mc-ps

Comments URL: https://news.ycombinator.com/item?id=46142840

Points: 2

# Comments: 0

New comment by chw9e in "Launch HN: Phind 3 (YC S22) – Every answer is a mini-app"

chw9e — Thu, 04 Dec 2025 01:40:04 +0000

As someone who never used Phind, how was it better than copilot? Did it integrate data from broader sources or just do a better job of presenting it? Was it faster?

Bug Fixing Is an ETL Problem

chw9e — Mon, 17 Nov 2025 18:19:58 +0000

Article URL: https://qckfx.com/blog/bug-fixing-is-an-etl-problem

Comments URL: https://news.ycombinator.com/item?id=45956297

Points: 1

# Comments: 0

Ask HN: How do you handle oncall at 15-30 engineers?

chw9e — Wed, 29 Oct 2025 16:54:31 +0000

I'm doing research on how oncall breaks down as engineering teams scale, specifically at Series A/B stage companies (roughly 15-30 engineers).

The hypothesis I'm exploring: oncall workflows were designed for small teams where everyone knows the codebase. They break at scale because of specialization and context loss before volume becomes a problem.

Some patterns I've heard so far:

- Oncall engineers spending 10-20% of time just routing bugs to the right owner

- Bug reports from CS/sales teams missing critical context (no logs, vague repro steps)

- "Couple hours" average per bug, mostly on investigation not fixing

- Session replay tools too expensive to run at meaningful coverage

I'm trying to figure out:

- Is this universal, or just specific to certain tech stacks/org structures?

- What's the actual breaking point - team size, user scale, something else?

- Has anyone solved this well? What worked?

If you're currently running oncall at a growing company, I'd love to hear:

- What percentage of oncall time is triage/routing vs. actual fixing?

- What broke first as you scaled?