I don't know why I never tried this in the past! SSHing my machine and vice-versa!

Just figured out that I could use my computer's terminal to send to my android's clipboard via SSH.

Pair it with Tailscale and we have a beast!

I haven't had to tweak an OS like Win 11 ever.

I learned from a friend to use Zod to check for process.env. I refined it a bit and got:

```

const EnvSchema = z.object({

  NODE_ENV: z.enum(['development', 'production', 'staging']),

  DATABASE_URL: z.string(),

  POSTHOG_KEY: z.string(),

export type AlertDownEnv = z.infer;

export function getEnvironments(env: Record): AlertDownEnv { return EnvSchema.parse(env); }

```

Then you can:

```

const env = getEnvironments(process.env);

```

`env` will be fully typed!

Definitely, I need to do some improvements in my frontend logic!

I'm in a process where application-level programming isn't cutting it anymore (I still have a lot to learn, but it's in the diminishing returns).

I've been looking to understand the entire stack at a deeper level (from how requests are made to how they're parsed), and this seems like the next natural step!

Thanks a bunch!

How did they manage to grow so rapidly in just 5 years and exit with a $32 billion acquisition from Google?

Comments URL: https://news.ycombinator.com/item?id=45832730

Points: 2

# Comments: 0

I love to have thumbnail tabs!

https://medium.com/@tomer2138/how-canaries-stop-prompt-injec...

Here, Derek DeJonghe, provides production-ready Nginx configurations with explanations for the most common cases.

Comments URL: https://news.ycombinator.com/item?id=44692274

Points: 5

# Comments: 1

Comments URL: https://news.ycombinator.com/item?id=43818498

Points: 2

# Comments: 0

I’m Jose, the solo founder of AlertDown, an automated user flow monitoring tool for web apps. It notifies you when something breaks—before your users do.

Imagine end-to-end tests with real-time monitoring, catching regressions and silent failures without:

• Writing a single line of code • Installing any dependencies

Just add your URL, and AlertDown will: 1. Extract all possible user actions (clicks, inputs, dropdowns, etc.) 2. Automatically test each flow, branching out dynamically 3. Detect silent issues like:

• Missing dropdowns • Unresponsive buttons • Third-party API failures

Unlike tools like Sentry or Datadog, these errors often don’t show up as obvious failures, which can lead to lost revenue.

Why I built this:

While working at a cash-constrained startup (not mine), we lost $1,347 in revenue due to a misconfigured feature flag at step 7 of our onboarding. We didn’t realize it until 3 days later—after a user reported it. :(

I’ve seen this issue repeatedly over the years:

• “Non-breaking” code silently causing regressions • Third-party services failing unexpectedly • Testing fatigue—running the same flows over and over manually

I thought an automated solution like this wasn’t possible, but after an initial POC and 5 months of work, it’s finally live and working!

Tech Stack: • Remix, TypeScript, Vite • Supabase • Docker on Google Cloud Run & Compute Engine Temporal.io for orchestration

What’s Next:

I’d love your feedback and ideas as I continue improving: • Handling login screens • Slack integrations for alerts • Custom viewport testing

Try it out now (for free)! I'm looking for some early adopters that would like to pilot the product.

You can access without paying (just head to the login page)—it’s currently in public testing!

Looking forward to your thoughts and feedback.

P.S: And if you are part of a company or building your SaaS, I want to work close with you to craft a unique experience.

Comments URL: https://news.ycombinator.com/item?id=42779844

Points: 1

# Comments: 0

And I think the latter is good enough for us to do exciting things.

What should I use for a startup going after B2B sales?

ULIDs or CUID2s for primary keys?

I remember seeing Hussein Nasser's Shopify video: https://www.youtube.com/watch?v=f53-Iw_5ucA

And it made me switch to using ULIDs instead.

I know this may fall into the "premature optimization is the root of all evils" category. But since it's straightforward, what should I do?

I've been using ULIDs in distributed systems with no problems so far.

Comments URL: https://news.ycombinator.com/item?id=41537558

Points: 2

# Comments: 1