abatilo

Speaking for myself, the language itself is simple. There aren't a lot of features, which means you do have to write more and it's more verbose, but learning the entirety of the language's features is very easy to do.

Even with the relative magic of things like init functions that execute on import, there's so much less magic that you have to consider.

IMO it's the easiest to jump into for this reason. Yes, the code itself might be excessive, but reading Go is as simple as can be.

Those bandwidth costs are going to be so expensive

In 2017, I worked at a company that built a websocket based app and ran it on Heroku. We had this exact same problem and all 5 of the engineers at the time spent all day until we realized that we had socket connections still connected to some "ghost dyno".

I can't remember exactly what we did but I think we just kept deploying over and over again until the old one finally went away.

I wonder how much of it is legitimately just because this stuff is insanely expensive to run, and they needed the money?

There's https://www.ethicalads.io/ if you're only targeting developers

I've had a few chats with Kyle and Jacob over the last few months.

They're incredibly knowledgeable about the subject and are making amazing strides for build speeds. I'd encourage anyone who doesn't believe these results and benchmarks to just try it out. They're completely real and it's delightful.

Used it in Kubernetes. When there's a breaking API change, imagine how terrifying it is and has been to try to upgrade everything in the Istio control plane when possibly your entire cluster's communication, observability, security, etc are all going through the same system. It became less stressful to do all the planning to build a second Kubernetes cluster and switch over with the new version already installed.

Hi! Is your email domain actually lambdal.com or should that be lambdalabs.com?

Pssst, your post says growing to 30 in 2022.

It's not sexy sounding but I just had to implement something similar, in house, for some BI workloads involving HR data for our people team. I had to ensure that I nor any other infra person would be able to backdoor and view any of the data or decrypt anything. I think this is another space where this could be very useful.

I interviewed with Claypot last summer for an infrastructure role. They responded about 5 weeks after I applied. My first conversation was with Chip herself, where she appeared to be checking Slack for our entire conversation.

I did a technical interview where the interviewer asked me to do calculate some statistics over streams in a Jupyter notebook. I completely bombed the interview, let that be clear. But it was very odd to me that for an infra engineer role, at no point was I asked about any experience with ML infra, scaling, operationalizing models, storage, networking, etc.

Nice overview of options

Copilot is calling out to OpenAI's Codex, so they don't really have anything on their own

Hi all, I recently slapped together https://terragen.sh/ in a single sitting. Terragen.sh is a website where you can put make some decisions about your infrastructure needs, and generate the terraform that matches your configuration.

It's an extremely narrow use-case right now as I put it together in about 4 hours, but it has a fully functioning extensible backend that uses versioned Go templates to store and render the terraform. Right now, the site generates terraform for the hello-world app for AWS AppRunner.

In an effort to get early feedback, I decided to post the link even though the website is still extremely limited.

Do you think there's a use case for a website like this? Maybe advanced configurations require payment to generate? Would you ever use a tool like this? Maybe if it also created best practices VPCs, database configurations, etc?

But also, maybe there isn't much of a market for anything more advanced!

Thanks for any and all feedback!

Good luck with your product! Hope you have some great success

Out of facilities

I have $2 MRR

I built https://catstories.ai

The UX is pretty terrible, honestly. But I thought I could get away with it since I didn't expect anyone to sign up.

I also built https://sudokurace.io based on an idea my wife and I shared while talking. We talked about it for years and I finally decided to build it, only for us to play it for a day or two and now we never really touch it.

Just my opinion, but I would be shocked to find out that people were trying to provision anything in their AWS accounts from their phone.

Reminds me lot early pramp.com! Very cool

Absolutely!

It's a typescript, react, tailwindcss frontend.

Golang backend with gorilla websockets. To coordinate which replica had the right websocket, the replicas communicated over a NATS server. State is backed by Postgres.

Hosted on my side projects' AWS EKS cluster!

Noted, thank you!

I switched from Evernote to obsidian.md last year and am very happy with my purchase. You have to pay for the ability to sync between devices but I'm okay with that.

Oh, neat.

Hi everyone. I made this website a few months ago. If there are questions or suggestions, happy to answer them.

The website was covered in the wallstreet journal and a few other places: https://www.wsj.com/articles/many-companies-want-remote-work...

Definitely overkill for the website itself but I also built it as a learning opportunity so it's a react frontend with Tailwindcss, Golang + gRPC gateway, hosted on an EKS cluster and backed by serverless Postgres Aurora. I hope the stack holds up with eyes from HN :D

There's at least some precedent for conditional execution in standard libraries.

At least, in Python, the default sort is https://en.m.wikipedia.org/wiki/Timsort

Not having any turn over in 6 years in our field is kind of amazing. Do you think you're doing something different than the standard set of practices? How have you retained people for so long?

Couldn't you just use something like Heroku then?

Then a company should raise their upper bound to lower the risk of that happening.

Wasn't the ability to add a //go:generate comment enough to make it easy to re-generate anything that you needed?

These visualization tools are so creative, IMO.

I had interviewed with https://www.streamlit.io/ and it seems that these kinds of tools can really make your data science teams super productive. It becomes so much easier for them to explain the impact of work that they're doing.