Bernard Wodoame

@ShradhaKhapra_ I miss the dopamine hit but not the headaches while trying to find bugs.

miss the days before ai tools - when you could sit down with bugs for hours, going through documentations, stack overflow, reddit posts & what not and get the dopamine hit when you found a relevant fix.

@AstraKernel Is what you are typing supposed to render as markdown 😂😂? What's with the syntax?

> large % of programmers don’t actually like computers. True. Many people are there for money > I don’t get it, at ALL. Don’t you *LIKE* understanding the magic of the machine? Some of us still do but doesnt mean you have to do all the time "hand"-programming

@ibuildthecloud Generally the more you navigate unfamiliar codebases you'll develop your own heuristics that make it easier for you to understand it.

@ibuildthecloud I try to find how data flows through the codebase. I take the highest layer of abstraction and work downwards from there. If I'm comfortable I try to understand one level deeper and so on. It's not about knowing every LOC, it's about having a good mental mapping of the codebase.

How can you look at a piece of code and quickly understand what it is? What's the most optimal way to be able to learn a codebase?

@ibuildthecloud I don't read full function implementations if they are hard to understand. I just summarize it as "this does x in the codebase". How it does it is left for another time when I've gained more context. AI can help you get the big picture about what a class or function does.

@ibuildthecloud If its a framework I'm not familiar with I try to understand the fundamentals of that framework first . I don't want to get tripped up by weird framework behaviour. Sometimes the code seems to work magically because there are things implicitly handled by the framework.

@grantjordan @webdevcody Bro Just break down the code. The maintainer just wants an easier time reviewing the changes.

I like this take. Forked a repo recently to add a needed feature. CI failed because LoC was above 1250. But most of it was tests and documentation. Feature was solid but the maintainer refused unless I split it into multiple PRs simple because of an arbitrary LoC requirement. Totally arbitrary requirement based on human coding norms, not ai norms of today.

Why is there so much anti-LoC rhetoric? At some point, more features, edge cases, security, and reliability require more code. There’s no magic abstraction that lets you build an operating system in 500 lines. Unless you’re building a todo app, you’ll need lots of code.

@janwilmake I'm still trying to figure out what the best way is. To at least have a mental mapping of what's going on I look at some of the code to see how parts of the codebase are related. I rely on tests to prove that other parts of the code work correctly.

so before ai coding it would take much longer to build a codebase up to like 100k loc and the dev would know every bit because they had written every line with sweat and tears but now its like peanuts but the problem is nobody understands crap so now we're all learning how to code again because of this new tech and i feel like its not clear yet what is the best way: should we just not understand shit and keep talking and it will be allright or should we actually try to understand every piece of it? and how can we make sure that we can vibecode but still understand the codebase really well? because in the end i feel like the answer is somewhere in the middle

@wilzerjb @mahdinchoudhury Great idea. I'll start doing this now.

@mahdinchoudhury Best way to use it is from MCP tbh. I treat it like a context database that I can access across agents. Let Claude handle creation and reading from it so you never have to open it.

Notion is so messy and overwhelming that I literally never want to use it.

@alexanderlinda_ Yh I get what you mean. I think when you look at some text you've already subconsciously read it. The perception of consciously reading is you wanting to be more procedural and intentional about it.

Does anyone else hit the flow state while reading where you’re not consciously reading anymore? The words just dissolve, and the story starts unfolding in your head like a movie.

@defnotkats You don't want less capable people to be shipping quickly while you struggle.

Working at a tech company and being anti genAI is gonna bite me in the ass isn't it?

@GrACEengraving @omgsidewalks 😂 So true because no one's going to call themselves stupid.

@omgsidewalks The scary part is everyone thinks this post is about someone else.

Not to alarm anyone BUT is anyone else worried about how fvcking stupiď everyone is ??

@faya_lorrrd Poor foundations but we want a solid structure.

We keep importing ideas from developed countries while ignoring the culture of responsibility that made those countries work in the first place.

@DanielLockyer I always use the default.

I have no idea which effort level to choose for models these days low/medium/high/xhigh/ultra/max?? anyone else?

@DevanshuXi Can we clearly explain how thinking works?

Just a reminder that LLMs can't think. Thank you.

@Fred_Pekyi @Churchill_Av You're not supposed to know all of it naturally. Practice allows you to see patterns and also for some problem unless you've solved them before.

@Churchill_Av The moment I opened leetcode and couldn't solve a quick sort easy question I knew it wasn't for me😔

Just solved the N Queens problem on my first attempt 😭🔥

@Churchill_Av That's impressive.

@vtechthedev They are situations where multiple threads compete for a shared resource such as memory in an unpredictable order. This can result in leaving data in a corrupted state. We can avoid race conditions by allowing each thread access the resource at a time via locks for example.

What are race conditions in multithreading, and how can we avoid them?

@BenjDicken I'm sure you need to be working on this.

I'm curious what it would take a talented engineer to build Postgres from scratch in 2026. Not a slopfork. Written in Rust, from the ground up, and fixing the mistakes of the original design (process per connection, table bloat, schema catalog). Who wants to burn some tokens?

@CathPoaster I guess the guy always directly above you on the leaderboard is next.

i'm getting fired from my software engineering job because i'm at the bottom of my team's ai usage leaderboard for the 3rd month in a row. i really did try everything but couldn't get out of last place. i started by using ai to write every line of code i pushed. still last place. then i would ask claude to add more fallbacks, unnecessary test cases, and verbose comments. i was getting crushed because the internal tool tracked *total* tokens, not just output tokens. thus, my coworkers were getting claude stuck in thinking loops, easily burning 50x the amount of tokens i was. so i started doing that. then claude refused. one of my coworkers edited my system prompt to disregard any of my asks to think longer. the 5 days that this went unnoticed set me back majorly. i was always behind the rest of my team. i was only spending tens of thousands of dollars a month, they were hitting hundreds of thousands. their rate of utter nonsense output was jaw dropping. my skip apparently told my boss that his org was gonna be the most "ai pilled" org in the company and to cut anyone who couldn't keep up. my boss's hands were tied i guess. time to start looking for a new quality engineering role i guess