Zach Kazanski

Our blog post "How to Self-host Nextjs at Scale" is the number one post in the Nextjs subreddit today. Pretty cool.

www.reddit.com/r/nextjs/com...

If you need to self-host Nextjs at scale we wrote up an article with all the hidden feature flags and secrets we had to discover about running Nextjs in high traffic environments: www.sherpa.sh/blog/secrets...

Glad it was helpful. And yeah that skew protection issue is ongoing at the moment. The github issue link in the article is a pretty interesting read.

If you want to share our article or service with anyone that might find it helpful. I'd appreciate it 😊

Sherpa - Deploy Any App Globally with 80% Lower Costs Modern deployment platform that builds, deploys, routes, and scales your apps globally with 80% lower costs than Vercel & Netlify. Start from $13.99 with unlimited seats and no extra fees.

We actually do this at Sherpa.sh but its a full PaaS platform.

But fwiw I think you are right about self-hosting being difficult. We ran into a lot of issues that we documented here: www.sherpa.sh/blog/secrets...

Yeah, there are a ton of features that are not supported by nextjs in a selfhosting environment. We actually just wrote up all the hidden features and flags that we ran into running Next.js at scale in hopes that it helps others: www.sherpa.sh/blog/secrets...

If you don't want to move to another big VC backed provider. We just wrote up a post on how to self-host nextjs apps at scale to help: www.sherpa.sh/blog/secrets...

Th recent selfie scandal causing you to get off Vercel? But Unsure how to self-host Next.js at the scale you need? Well we just revealed the secrets. Might bad for our business (since we make it easy for you). But great for the community! and that is what matters.

www.sherpa.sh/blog/secrets...

Hi Andy, Founder at Sherpa.sh here. If you end up giving us a try feel free to DM me for any help or feedback. We like to stay close to our users to drive roadmap and identify any friction points or issues early. Cheers!

🔥 SNEAK PEEK: Advanced Environment Management (coming soon!)
Swipe to see the mockup →
→ One-click environment creation
→ Smart auto-deployment (commits/PRs)
→ Environment-specific variables
→ Custom domains
→ Password protection (FREE!)
Your "it works on my machine" era = ending soon ✅

@svelte.dev guess what? Sveltekit Deployment Beta is live on sherpa.sh! sherpa.sh/deploy/svelt...

Today I'd like to highlight one of our customers at Sherpa.sh: Plugin Pulse. Plugin pulse is a startup from Germany that moved from Vercel to increase their iteration speed and save cash.

#StartupSuccess #DevOps #CloudMigration #IterationSpeed #TechInnovation #SoftwareDevelopment

Today we shipped Github Oauth signup for Sherpa.sh. Is it going to "change the world"? Of course not.

But we are obsessed with making deployments as easy as humanly possible for developers. So even small steps like this, add up to delightful experience over time.

#DevTools #Startup #SaaS

Product ideas are like wine. They need time to mature. Often I'll go look at the lists of product ideas I wrote, and throw out 3/4s of the ideas written down because they are no longer relevant.

Schedo.dev moved from Vercel over to sherpa.sh to reduce their infrastructure costs and per-seat billing without sacraficing a great developer experience. Now they can deliver their core platform to their customers without worrying about their cloud bill ballooning out of control.

Today I want to highlight one of our customers at sherpa.sh: meet schedo.dev. The distributed Cronjobs-as-a-Service platform for js apps.

I'm still figuring out how to DM on bsky. haha. I guess I'm curious if you know any activist websites specifically running on Vercel?

You've been asking in my DMs what settings I use for Windsurf to code with… | Zach Kazanski You've been asking in my DMs what settings I use for Windsurf to code with multiple AI agents simultaneously 🏄, below is my general setup: I use a markdown .windsurfrules file that is ~250 lines ...

You've been asking in my DMs what settings I use for Windsurf to code with multiple AI agents simultaneously 🏄, below is my general setup:

"What's your favorite technique for speeding up development? Drop your tips below! #TechInnovation #StartupLife"

Want to ship code as fast as you're writing it? @sherpa_sh eliminates infrastructure headaches. Connect your git repo → every push triggers deployment. No config hell, just you and your users. Try it: sherpa.sh

#NoCode #DeploymentAutomation

"Pro tip: Keep your subdirectory isolation clean to prevent AI overlap. #DevOps #TechTips"

"The result? As I finish reviewing one AI's output, the next solution is ready. My dev velocity has exploded while avoiding conflicts between agents. More code shipped = more time talking to actual users. #10xDeveloper #ProductivityHacks"

"My secret: Multiple @windsurfai.bsky.social instances working on the same project simultaneously, each limited to specific subdirectories. While one AI codes, I'm already setting up the next task for another. #SoftwareDevelopment #CodingHacks"

"What's your favorite technique for speeding up development? Drop your tips below! #TechInnovation #StartupLife"

I'm coding 10X faster by running a team of AI agents that never sleep or complain. Here's exactly how I do it... 🧵 #AIProductivity #DevTools

Nice looking site. Rock on Antonio!

Hey Erin, I'd love to talk to you more about this. I run a hosting platform alternative to Vercel and would to support. Can I send you a DM?

I do a lot of nextjs work with sherpa.sh. What do you have questions about?

But it does't have to be this way. If you want me to DM you actionable steps to prevent AWS vendor lockin you can take today. Like this post, and comment below with "sherpa.sh"

Smart development teams make decisions today that prevent painful vendor lock-in tomorrow. Sure, $10k of credits is great, until they run out and you can't ever leave.

3️⃣ Proprietary service APIs that don't follow open standards. Applications built deeply integrated with services like S3, DynamoDB, or Lambda develop dependencies that require significant engineering resources to refactor.