Great code Overviews

• Architecture-Infra codes

• roadmap.sh

• interview w JS senior developer -

  • 

• stacks

  • Fireship overengineered stack 2021
  • 

• Already in place

  • web dev stages.excalidraw
  • CMS problems & solutions - from builder.io
    • what? can find this info in its specific notes
  • fullstack “definition” & roles/pipeline -
    • Product manager - manager + product strat
      • acts as a manager and has some business tasks related to the product success mostly vision and purpose of it
    • i
      • 
  • Frameworks
    • batteries included vs not v
      • quick list
        • 
      • not - need to choose every piece yourself
        • complexity scales with needs
        • could end up with unmantained garbage
      • included - there’s a standard solution & not much more
        • can be complex from the start
        • could need to modify it to fit your needs
        • usually follows the MVC mentality
          • realistically it gets out of hand at a certain point and it should not be called MVC anymore
      • i
        • 
        • 
    • which framework?
      • t3 stack?
      • App or Website?
        • 
      • Ease of use vs simplicity
      • backend&frontend
        • htmx goes from backend to a little frontend
        • next.js goes from frontend to a little backend
  • Learning VIM be like v
    • i
      • 
  • Backend
    • which db? K/V > Document > Relational
      • Try to mentally start w K/V first then if you need more stuff just go with relational
      • 
    • Serverless, Traditional servers, & fluid compute Theo v
      • traditional dedicated servers give 1 entire thread
        • u pay for these ~2.5 seconds
          • 
      • serverless - pay based on compute time
        • properties
          • stateless (people use cache/DBs to have it)
          • spin up & scale on-demand
          • die when not used
          • 1 instance can serve only 1 user at the time
        • spawns a VM per-request
        • u pay based on compute time, not threads
          • paying for 6 whole seconds
          • the goal is to reduce the compute time in milliseconds
          • 
        • doesn’t support streaming
        • waitUntil
          • the lambda don’t die instantly but can do more even after the “final” response
            • thanks to vercel abstraction of the AWS gateway (vercel edge)
        • announced - in-function concurrency
        • vercel edge checks usage of an alive lambda and if it’s not too much it gives it another task
        • could also try to understand if other devs have tasks that can fit in there
        • you still pay for the whole time even if you might be able to do more with the same amount of time
      • Pay based on CPU execution time
        • cloudflare worker
          • gives request to a v8 instance
          • pay based on CPU execution time -
          • no compute time limit
          • bad stuff about this -
            • it’s not node
        • fluid compute
          • why not enabled by default -
            • because some code expect multiple requests per-task
          • for AI chat apps
            • most of time is spent waiting to then stream a word/token and time of compute gets very big
          • works like cloudflare workers
          • practical benefits -
    • myIT-Infra
      • workflow overview
        • 
  • Dev cody
    • next.js fast explanation -
    • Gitflows/ship like companies - gitflow similar trunk based
      • gitflow
        • 
      • trunk
        • 
    • Mindmap of features
      • 
      • 
      • 
      • 
      • 
  • devops
    • workflow overview
      • 
  • Waterfall vs Agile -
    •