@tom_doerr Good one! You have others similar to Metabase in your invetory? Looking for something code based to develop the views

@yacineMTB heya kache, my pleasure

Technical debt is now a choice, not a necessity. With GenAI-powered refactoring, there's no excuse for messy code anymore. The speed vs. quality trade-off is dead. Clean architecture is now a continuous, low-cost process. The future is here.

@tdinh_me @nextjs @render Do you leverage a tool like Sentry? You probably could get deeper insights what‘s going wrong.

@thesamparr I use a simple approach with Cursor and memory files (simple text files). When working on something I from time to time let the LLM summarize the learnings and write them in a file. Then later, if the same context is required, I reference it in the chat again and proceed.

@tokenizeitHQ Interessanter Case 👀

one day this will be part of an elaborate exploit chain

@paul_dentro @droosevelt @pipedream What are you currently using Paul, to power your deployables?

@paul_dentro @droosevelt @pipedream Cursor as IDE with Sonnet 3.5 to build Node.js scripts and Pipedream as infrastructure foundation to deploy automation workflows / pipelines is bonkers. 🤙🏼😬

@petergyang You learn way faster, than when coding without AI, cause the feedback loop is shorter. You always can ask for context specific explanations, without having to dig through tons of docs. You faster get an understanding how things might interact. Architecture is key :)

@levelsio Why not use an UI, which works for multiple models like TypingMind? Would not want to trade the model capabilities of Claude when coding for some UI.

@DhravyaShah Spin up some metal in other locations than Germany?

@yacineMTB It‘s always has been about solving problems for users. Devs who can‘t work autonomous in their respective scope are not worth the money.

@yacineMTB haven‘t used Google for a while now, besides looking for places to go. Do you know about TypingMind? You can chat with a bunch of models and integrate Perplexity as plugin right in the chat

@deepfates Dunning Kruger helloooo

@allenholub @conways_law Thank you Allen!

@conways_law @allenholub How would one sensible differentiate between OO and Domain Driven Design? For me they both share the hollistic view on a problem domain with the goal of implementing a sensible abstraction.

@dashboardlim Clients

@TrainedOnTest Gosh I love it. If you capture the call graph of the invoked code, you also could derive high risk areas and uncalled code. How are you analyzing the repo?