Ask HN: Solo-devs what is your tech stack?

I've worked on several solo projects in the past. I've personally found the below stack to be quite productive:

Frontend: NextJS + Tailwind Backend: Flask Ops: Heroku + addons

Express.js + React + Postgres

For hosting recently started using Render which is even faster to setup than Heroku, 5 minutes from signup and I had everything setup with ssl and custom domains.

Boring, but functional:

Flask, MySQL, Bootstrap, jQuery

Node.js/AWS lambda/DynamoDB/Kinesis for backend

Pytorch/k8s/AWS fargate for Deep learning api

Clingo/Prolog for symbolic AI

React.js/flexbox/sass for frontend

SwiftUI for iOS

web dev: nextjs + scss + contentful cms + netlify

Next.js + GraphQL + Postgres

Written in TypeScript.

Rails + Rails Jumpstart Pro[1] + Hatchbox (& AWS)[2] + Tailwind UI

[1] https://jumpstartrails.com/

[2] https://www.hatchbox.io/

[3] https://tailwindui.com/

Lazarus (Free Pascal) / MySQL

Scala (zio, http4s, circe, doobie), postgres, scala.js (react, slinky, bootstrap)

Haskell (Warp HTTP server w/ WAI middleware, Selda with SQLite or Postgresql, blaze-html for templating).

I try and avoid JavaScript but if I use it I usually write vanilla JS and avoid package managers and build pipelines.

Swift, SwiftUI, UIKit/AppKit for iOS, iPad and mac

FastAPI + Postgres backend

React + TS for web

Edit: some pretty exotic stacks in here! Cool to see.

ruby vuejs + vuefity + postgres + heroku

Rails, sprinkles of React, Bulma, Postgres, Redis, and Hatchbox on Digital Ocean.

uWeebSockets.js: better performance and api than expressjs, hapi, koa, fastify

React: stable, doesn't break, probably lindy.

Tailwind css: it just works.

Postgresql, elasticsearch, clickhouse, redis.

Namecheap, digitalocean.

For backend: Golang + SQLite + Nginx For frontend: Nuxt.js / Vue.js using the SSG option

I meld the two using the JAMstack design in Nginx where the backend is exposed on /api path and the static site is hosted from the dist directory on the root path of the site. And while I use Go in production, I typically use Python for making proof of concepts or writing any quick glue scripts.