more validation that leaving Serverless Framework was the right move. I suddenly am unable to deploy changes to one of our other stacks. no useful error, just “Access Denied”. 😕 the funny thing is, nothing changed, to my knowledge. it just stopped working today

with SST how does everyone define all of their API Paths? I was interviewing with a company and in their technical challenge they insisted the API paths be defined via an OpenAPI file. I had not used an OpenAPI file since moving to CDK, I used it for one project with AWS SAM. I used SST for my solution, I found using

apigateway.SpecRestApi

to be limiting. I don't think the issue was AWS CDK, I think the issue is insisting on an OpenAPI file. So I'm just curious what do most companies use. Do many companies really use OpenAPI files?

It looks like @Akos ‘ post is on HN!

^ yeah the way I think about it is the decision is pretty simple 1. If you want to run everything locally, you have to pick open source tools that can be scaled down into a container (RMQ, Redis, Postgres, etc) 2. If you want to take advantage of very purpose built closed source managed services (sqs, eventbus, snowflake, etc) you need to develop in the cloud (and find a way to make that smooth) With #1 I actually think local development sucks. You have to use something like docker-compose to bring stuff up, stuff never works right across different os's people have, there's .env files you need to pass around. And I also think it's weird to decide what you're using in prod based on how you want to develop things With #2 there are rough edges but it was a lot worse even a year ago so things are improving quickly

I really wanted to just leave a comment there being like, "none of this matters - just go to plain.com and look at what they're shipping and are you really going to tell me they're a disaster"

GraphQL people do you like this - https://github.com/hayes/pothos Basically lets you define the gql schemas + resolvers together. Similar to nexus but this gives you 100% typesafety without any codegen on the backend

Hi folks, This is Serkan. I am CTO & Co-Founder of Thundra (thundra.io) which is tracing and debugging solution for serverless applications. In my free times I am working on some side projects and last weekend I have implemented a simple utility (

treq

) to trace module imports/requires to show you the most expensive dependencies to be imported/required during coldstart: https://github.com/serkan-ozal/treq I hope you find it useful and it help you while investigating coldstart delays. Waiting for your feedbacks. Cheers -- Serkan

Oh, cool

I’m very pleased to announce @Gabriel Araújo has joined Leadent Digital to help us build out our serverless products on Serverless Stack 😄 🎉

This is what I go to Twitter for https://twitter.com/thdxr/status/1510814420691083267?s=20&t=I1fJzUyxQdfZMjA6vG5Dsw

lol

What are some non-AWS services you guys are using? Looking for the top few you'd almost always use no matter the project - eg Sentry

Stripe and XState. Something for feature flags

Hi folks, I have launched my hobby project Globadge (https://globadge.com): a free badge service. And the first badge goes to serverless of course :) You can get latest version of AWS Lambda layer as badge https://www.globadge.com/badges/serverless#aws-lambda-latest-layer-version Have an idea for a new badge? Please let me know it :)

If you're just laying around today at 4pm CDT and wanna hear be ramble on about Serverless come on by my talk 🙂 Will be showing some SST and Arc.codes. Pretty basic stuff, but should be a good time, or maybe not who knows. https://twitter.com/rawss/status/1513953621335023619

Thoughts on this hook? Yay or nay? 😄 https://gist.github.com/spion/09af04ea2ae8b6858e8edce11002b063

So I've just discovered CDK Custom Resources and they seem pretty cool. It got me wondering, has anyone used this before for more interesting use cases? I see that you can run aws-cli commands with it (e.g. validate an SES domain), but I was wondering if people used it for stuff outside AWS, I can see it useful for creating resources outside AWS and use the results inside cdk.

Howdy y'all! My company is planning an AWS/SST ML event (may LIVE stream) at Capital Factory in Austin, TX. Would any of you be interested in attending a LIVE stream 👀? What other topics would you like to see covered, as far as machine learning and SST go? Here's the tentative description: "Build, Train and Deploy a Model on AWS: This event will cover end-to-end use of Amazon Web Services to build, train and deploy an accurate machine learning model using SaaS product for example. You'll learn how to leverage Serverless Stack (SST), AWS Lambda, S3, Sagemaker, Amazon DynamoDB to create a highly available, scalable and affordable machine learning model in production."

Built with sst: https://www.producthunt.com/posts/tailwindpdf Leave a feedback/upvote if you find this useful 🙏

Also @thdxr makes his own memes https://twitter.com/thdxr/status/1516394798407692291

Any recommendations for higher level TS interfaces to dynamodb than DocumentClient?

RIP node-fetch ⚰️ https://nodejs.org/en/blog/announcements/v18-release-announce/

Woooah

does anyone here use vscode breakpoint debugging regularly and see issues with breakpoints being missed or high cpu usage?

For those of you following domain driven design patterns - have you ever considered using GQL as the domain interface given its ability to do fairly complex things? I'm noticing my domain and GQL api tend to be 1:1 for the most part and the pure domain layer is starting to feel like boilerplate. A lot become simplified if you collapse them into one

curious if people prefix their IDs in their database or API, eg

usr-<ulid/uuid>