Debugging Like a Senior: Console & Network Tab Techniques

I wasted hours debugging the wrong part of my system. It was maddening. Production was down. I'd jump straight to the last error message, then deep-dive logs, pulling my hair out. Turns out, I was looking in the wrong place every single time. The "fix" wasn't a new tool. It was just changing how I *read* error messages. Errors are not just a red line. They're a map. → The first error in a chain is usually the root cause. → Stack traces aren't random. Find your own file/line number in the trace. That's where the problem *originated* in your code. → Don't just read the top. Trace the execution path. This simple shift changed my debugging game entirely. What’s the small shift in your debugging approach that made the biggest difference? #SoftwareDevelopment #DebuggingTips #DeveloperProductivity

Claude Code won't stop. Spent 30 minutes yesterday watching it obsess over dependency versions. Meanwhile, the library I needed had a skill sitting right there. Ready to go. Did it use it? Barely. I gave it a task, it started strong, then suddenly it's deep in dependency hell trying to lock down versions that weren't even the point. The model isn't dumb. It just got stuck on the wrong thing. This is the harness problem. Everyone's talking about better tools, better memory, bigger context windows. But what about focus? Knowing when to stop digging and just use what's already there? Rabbit holes are real. And they’re a harness engineering problem, not a model intelligence one. What's the weirdest rabbit hole you've seen an agent go down? #AgenticAI #AIEngineering

Every developer knows the feeling: something works perfectly in your environment but fails elsewhere. Enter cache invalidation, the silent disruptor that can turn a smooth deployment into a debugging nightmare. This meme reminds us that while 'It works on my machine' is a common refrain, it’s not always the full story. Cache issues can lurk beneath the surface, affecting performance and user experience. Let’s embrace this as a reminder to test thoroughly across environments and consider cache management early in our development process. When cache invalidation joins your meeting—software's version of 'It works on my machine.' #DevLife #SoftwareDevelopment #CacheManagement #Debugging #TechMemes #EngineeringHumor

There are two ways to approach a bug. First: Try different fixes. Change things. Hope something works. Second: Pause. Understand the system. Trace the flow. Ask what actually changed. The first one feels faster. The second one works better. Because debugging is not about trying harder. It’s about seeing clearly. #softwareengineering #debugging

For quite some time, I used to support more than one rust async runtime in a single crate. After spend several hours troubleshooting which project has enabled two conflicting feature flags in my crate, I came up with few conclusions: 1. If you need support tokio, smol, monoio, compio or glomio as async runtimes, isolate parts which has calls to their APIs into separate crates, at debugging time you won’t have code paths being mistakenly activated 2. Type aliases are useful, but don’t use them to simple give another name to a struct, like File or RwLock, that will cause compilation errors if feature flags aren’t additive, but exclusive. In short, take a wide picture of your solution, if you codebase has parts which are clearly exclusive to one or another runtime, please save time and do not use a feature flag to support both of them, it will not only add code bloat, but verbose, hard to maintain and debug. #rust #programming #async #bestpratices #softwaredesign #featureflags

🐵 The Monkey Patch: Genius or Chaos? It’s the ultimate "break glass in case of emergency" move. Monkey patching—modifying code at runtime—can save your project or sink your team. The Reality: ✅ Pros: Instant hotfixes for buggy 3rd-party libraries and easier mocking in tests. ❌ Cons: It’s "invisible" code. It creates bugs that don’t show up in the source files, leading to massive debugging headaches. The Rule: If you must patch, document it heavily and make it your last resort, not your first. Is monkey patching a legitimate tool or just "duct tape" for bad architecture? 👇 #SoftwareEngineering #CleanCode #Programming #TechDebate

Last Fall, I added a stride scheduler to the xv6 operating system, to get a better understanding of scheduling disciplines. While testing, I found a condition that led to an apparent system hang at startup. I implemented an unsatisfying workaround at the time because I was focused on something a bit different. I decided I'd get back to it later. And, I finally did! A few days ago, I spent some time digging into the issue, and then I wrote up what I found in a blog post. Like a lot of fun problems, the fix was quite simple in the end. If you're interested in some of the interesting challenges of debugging code that doesn't follow the usual top-to-bottom execution flow of synchronous code, you might enjoy the post. https://lnkd.in/ejVaUxWa

Spent hours debugging a bug that made zero sense. A simple condition was failing: 4 / 2 == 2 Logically correct. Mathematically correct. Still… failing. No crashes. No obvious errors. Just a silent mismatch breaking the flow. Out of frustration, I printed the value instead of trusting it. 👉 1.999999995 That’s when things clicked. It wasn’t a logic bug. It was a precision problem. Floating-point arithmetic doesn’t guarantee exact values — numbers are stored as binary approximations. And sometimes, that “2” you trust is just… slightly off. That tiny difference was enough to fail an equality check: if (result == 2) → false The fix? ✔️ Avoid direct comparison of floating numbers ✔️ Use a tolerance (epsilon) for comparisons ✔️ Or switch to precise types where exact values matter That day debugging reminded me: The scariest bugs aren’t the ones that crash. They’re the ones that look perfectly correct. And sometimes… 4 / 2 isn’t 2. #SoftwareEngineering #Debugging #Programming #Flutter #Android #Bugs

I rendered the video below entirely from the command line. Stack: HyperFrames (our new open source framework) + HeyGen. Game changer. 🚀 💜 Agents and Humans can now make and edit videos like these:

HeyGen is absolute 🔥🔥🔥 People are really just starting to play across modalities like this but it’s powerful, actionable and easy.