Jacob Michelsen

If you think gold is expensive, think again. I've seen more companies and developers throw away their one precious asset than I can count. It happens. Every. Single. Day. That priceless asset is user trust. Here are the top 2 ways I see apps abuse it: Ask for permission when you don't need it. Ask for permission without explanation. "Allow app to take pictures and record video?" "Allow app to access this device's location?" Those are uncomfortable questions to get asked. If you abuse or obscure your intent, your users will quickly lose trust in you. User trust is your foundation. User trust is your north star. User trust is your income. Throw it away and watch your company tank. You'll get hammered at every turn. You'll get hammered on Google Play. You'll get hammered on the App Store. You never want to be a place where you need to ask forgiveness for asking for permission.

5

1 Kommentar

𝐌𝐢𝐜𝐫𝐨𝐬𝐞𝐫𝐯𝐢𝐜𝐞𝐬 𝐚𝐫𝐞 𝐫𝐮𝐢𝐧𝐢𝐧𝐠 𝐲𝐨𝐮𝐫 𝐩𝐫𝐨𝐣𝐞𝐜𝐭𝐬. Yes, I said it. We were sold the dream: 🔹 Independent services = faster deployments. 🔹 Distributed architecture = better scalability. 🔹 Microservices = the future. But here’s the dirty little secret no one talks about: 𝐘𝐨𝐮𝐫 𝐦𝐢𝐜𝐫𝐨𝐬𝐞𝐫𝐯𝐢𝐜𝐞𝐬 𝐚𝐫𝐞 𝐡𝐨𝐥𝐝𝐢𝐧𝐠 𝐲𝐨𝐮 𝐛𝐚𝐜𝐤. Why? Because: 🔸 Poorly defined service boundaries = constant debugging nightmares. 🔸 Communication overhead = slower systems and frustrated teams. 🔸 Scaling services independently = a myth when your architecture is tangled. Here’s the reality: 𝐌𝐢𝐜𝐫𝐨𝐬𝐞𝐫𝐯𝐢𝐜𝐞𝐬 𝐰𝐨𝐫𝐤 𝐨𝐧𝐥𝐲 𝐰𝐡𝐞𝐧 𝐲𝐨𝐮 𝐝𝐞𝐬𝐢𝐠𝐧 𝐭𝐡𝐞𝐦 𝐰𝐢𝐭𝐡 𝐜𝐥𝐚𝐫𝐢𝐭𝐲 𝐚𝐧𝐝 𝐝𝐢𝐬𝐜𝐢𝐩𝐥𝐢𝐧𝐞. Here’s how I help teams overcome these pitfalls: ✅ Establishing clean service boundaries that reduce interdependencies. ✅ Using AI-powered tools to optimize communication across services. ✅ Architecting for real scalability, not theoretical perfection. With years of experience in designing robust, scalable systems on Google Cloud Platform, I’ve seen it all— the good, the bad, and the spaghetti code. 𝐓𝐡𝐞 𝐠𝐨𝐨𝐝 𝐧𝐞𝐰𝐬? You can untangle the mess and build microservices that deliver on their promise. Let’s talk. Whether you need guidance on microservices, AI, or cloud architecture— or just want to brainstorm solutions—reach out. 𝐁𝐞𝐜𝐚𝐮𝐬𝐞 𝐲𝐨𝐮 𝐝𝐨𝐧’𝐭 𝐡𝐚𝐯𝐞 𝐭𝐢𝐦𝐞 𝐟𝐨𝐫 𝐚𝐫𝐜𝐡𝐢𝐭𝐞𝐜𝐭𝐮𝐫𝐚𝐥 𝐫𝐞𝐠𝐫𝐞𝐭𝐬. #Microservices #SeniorEngineers #GoogleCloud #SoftwareArchitecture #TechLeadership #SoftwareEngineering #ArtificialIntelligence #coding

3

Writing code is like a boxing match with a guaranteed knockout. Here is what I learned from two world champions. In one corner: clean, maintainable code. In the other corner: clever code. Clean code beats clever code. Any day. Any time. Every time. Clever code is hard to read. Clever code lacks context. Clever code strokes egos. It's usually just a power move from an engineer who wanted a moment to flex. But they often don't remember how it works 6 months later. And their colleagues never understand it from day 1. Don't pull a power move. It doesn't impress anyone. Clean code is self-documenting. Clean code follows a train of thought. Clean code can be read 6 months later. Clean code > clever code. The perfect knockout.

5

I don't want to be impressed by your code. Ever. I want to understand your code. Always. Clever code is confusing. Impressive code is unnecessary. Engineers who flex often later forget. If you really want to impress me and my team, here is how to do it: - write descriptive variable names - write descriptive function names - make them long, when necessary - break your code into multiple lines - make your code describe itself like a book I like my code to read like a novel, even if that means it's longer. I like confusing code to clarify itself by using variables to explain how it works in parts, one bit at a time. No more confusing logic. No more out-of-date comments. No more head scratching 6 months later. Now that is impressive code.

33

7 Kommentarer

𝗕𝗲𝗮𝘁 𝗜𝗺𝗽𝗼𝘀𝘁𝗼𝗿 𝗦𝘆𝗻𝗱𝗿𝗼𝗺𝗲 𝗮𝗻𝗱 𝗦𝗵𝗶𝗻𝗲 𝗮𝘀 𝗮 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗿! 🌟💻 Hey, developers! Let’s talk about something we’ve all felt at some point—𝗶𝗺𝗽𝗼𝘀𝘁𝗼𝗿 𝘀𝘆𝗻𝗱𝗿𝗼𝗺𝗲. That nagging voice in your head that says, “𝗜’𝗺 𝗻𝗼𝘁 𝗴𝗼𝗼𝗱 𝗲𝗻𝗼𝘂𝗴𝗵,” or “𝗜 𝗱𝗼𝗻’𝘁 𝗯𝗲𝗹𝗼𝗻𝗴 𝗵𝗲𝗿𝗲..” Guess what? 𝗬𝗼𝘂’𝗿𝗲 𝗻𝗼𝘁 𝗮𝗹𝗼𝗻𝗲. In fact, over 58% of tech professionals experience this at least once in their careers. It tends to creep in during the big moments: When you’re presenting your code to a client or team Starting a challenging new project Or even when you’re about to hit “deploy” on something you’ve worked so hard on Despite all your hard work, you might find yourself thinking, "𝗪𝗮𝘀 𝘁𝗵𝗮𝘁 𝗷𝘂𝘀𝘁 𝗹𝘂𝗰𝗸? 𝗗𝗶𝗱 𝗜 𝗿𝗲𝗮𝗹𝗹𝘆 𝗱𝗲𝘀𝗲𝗿𝘃𝗲 𝘁𝗵𝗮𝘁 𝘀𝘂𝗰𝗰𝗲𝘀𝘀?” Let’s get this straight: You’re here because of your skills, your dedication, and your growth—not luck. That voice of doubt? It doesn’t define you. 𝗪𝗵𝘆 𝗬𝗼𝘂 𝗦𝗵𝗼𝘂𝗹𝗱 𝗕𝗿𝗲𝗮𝗸 𝗙𝗿𝗲𝗲 💡 Impostor syndrome holds you back from celebrating your wins and fully realizing your potential. When you stay stuck in self-doubt, you miss the chance to embrace what makes you unique and valuable as a developer. It’s time to change the story you tell yourself. You deserve to feel proud of your achievements, no matter how big or small. 𝗛𝗼𝘄 𝘁𝗼 𝗙𝗶𝗴𝗵𝘁 𝗕𝗮𝗰𝗸 💪 Here are a few ways to silence that inner critic and embrace your confidence: 1️⃣ 𝗖𝗲𝗹𝗲𝗯𝗿𝗮𝘁𝗲 𝗣𝗿𝗼𝗴𝗿𝗲𝘀𝘀, 𝗡𝗼𝘁 𝗣𝗲𝗿𝗳𝗲𝗰𝘁𝗶𝗼𝗻: Did you learn something new? Fix a tricky bug? Ship a feature? Celebrate it! Each step forward, no matter how small, shows you’re growing and leveling up. 🎉 2️⃣ 𝗗𝗼𝗻’𝘁 𝗖𝗼𝗺𝗽𝗮𝗿𝗲, 𝗘𝗺𝗯𝗿𝗮𝗰𝗲 𝗬𝗼𝘂𝗿 𝗝𝗼𝘂𝗿𝗻𝗲𝘆: It’s easy to look at others and feel like you’re not enough. Remember, everyone’s path is unique. Focus on where you started and how far you’ve come. 🚀 3️⃣ 𝗕𝗲 𝗢𝗽𝗲𝗻 𝗔𝗯𝗼𝘂𝘁 𝗖𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲𝘀: Struggling doesn’t mean you’re not skilled—it means you’re learning! Talk to your peers, share your doubts, and you’ll find you’re not the only one feeling this way. 💬 4️⃣ 𝗦𝘁𝗮𝘆 𝗖𝘂𝗿𝗶𝗼𝘂𝘀, 𝗞𝗲𝗲𝗽 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴: Technology is always evolving, and no one knows everything. Learning something new, even if it’s small, is a sign of strength, not weakness. 🔍 Success isn’t about being perfect. It’s about showing up, learning, improving, and trying again. So, the next time impostor syndrome tries to hold you back, remind yourself: ✅ 𝗬𝗼𝘂’𝘃𝗲 𝗮𝗹𝗿𝗲𝗮𝗱𝘆 𝗮𝗰𝗵𝗶𝗲𝘃𝗲𝗱 𝘀𝗼 𝗺𝘂𝗰𝗵 ✅ 𝗬𝗼𝘂’𝗿𝗲 𝗰𝗮𝗽𝗮𝗯𝗹𝗲 𝗼𝗳 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝘄𝗵𝗮𝘁𝗲𝘃𝗲𝗿 𝗰𝗼𝗺𝗲𝘀 𝗻𝗲𝘅𝘁 ✅ 𝗬𝗼𝘂’𝗿𝗲 𝗽𝗮𝗿𝘁 𝗼𝗳 𝗮 𝗰𝗼𝗺𝗺𝘂𝗻𝗶𝘁𝘆 𝘁𝗵𝗮𝘁 𝘀𝘂𝗽𝗽𝗼𝗿𝘁𝘀 𝗮𝗻𝗱 𝗯𝗲𝗹𝗶𝗲𝘃𝗲𝘀 𝗶𝗻 𝘆𝗼𝘂 Drop a comment below—how do you deal with impostor syndrome? #DevMotivation #ImpostorSyndrome #YouBelongHere #Developers

2

#GitOps is fantastic. But it can be exceptionally bad depending on the the level of abatraction developers need to push to it and how easy you make it. On top of GitOps you could make a simple automation. So instead of dealing with git clone, add, commit, push.... Maybe just state your intent. "release this to environment x" --> enter. That's what we at VELUX do with #Bura. Bura release -v 1.2 -e dev Let us know if you want to know more about Bura in the comments.

32

18 Kommentarer

The new Qwen-2.5-Coder is 🔥🔥🔥 and this is how to test it and improve it for your use case Say you want a highly capable Rust coding assistant. You need to evaluate Qwen for this, improve the prompt, and potentially fine-tune it. So we’ve built this Colab to get you started, end-to-end: 1. Build a synthetic code dataset for Rust with distilabel. 2. Use this dataset to do human evaluation with Argilla. All on the Hugging Face Hub: open source tools and models FTW! Colab: https://lnkd.in/dxVFHvKD Next step: use correct responses as few-shot examples to make responses more accurate.

254

12 Kommentarer

Fredrik Lundh’s lambda Refactoring Recipe If you find a piece of code hard to understand because of a lambda, Fredrik Lundh suggests this refactoring procedure: 1. Write a comment explaining what the heck that lambda does. 2. Study the comment for a while, and think of a name that captures the essence of the comment. 3. Convert the lambda to a def statement, using that name. 4. Remove the comment. These steps are quoted from the “Functional Programming HOWTO”, a must read. source: Fluent Python by Luciano Ramalho #python #lambda #fluent_python

15

2 Kommentarer

Trying to get the right exception message for failed #powerautomate flows is anything but easy. I spun up a LowCode Plugin that did ok, but not quite everything. Inspired by Agnius Bartninkas xpath solution I made some tweaks and you guessed it, spun up a chrome extension for that 😎 Power DevBox Exception automatically creates the right expression for your particular flow, that way every message is caught with zero effort. Available from Chrome Store here :https://lnkd.in/e2Zc3TME (coming soon to Edge store but you can still add the Chrome store verision for now) All my other free extensions are here :https://lnkd.in/eefUbb2K (Chrome store again but you can they are in Edge store so you can just search for them there too) https://lnkd.in/eKuuN7Wt

55

5 Kommentarer

#OpenAI - side note on a new feature - but huge impact #OpenAI just recently announced: https://lnkd.in/eS8t-ri4 A "strict" structured output in JSON - oki, so what you might think, it was somewhat possible before - but: - it is a reliable output with a JSON format only (from a given question/text/etc ) - we can now use our schema -> which saves hours of defining interoperability with other program outputs, customers, etc .. just throw your data at the #LLM and get your data format back - whatever JSON format you want. - it might make endless standardization meetings obsolete whenever two systems want to communicate - we all had that: I need 8 numbers for my order numbers - oh no - we should use 12, etc .. and the re-programming effort -> now: all gone This new feature has the potential to speed up innovation and avoid super boring meetings :-) #AIControlledByYou #Innovation #SpeedUp #NeverStopLearning #GenAI

18

𝗔𝗹𝗹-𝗶𝗻-𝗼𝗻𝗲. 𝗜𝘀 𝗶𝘁 𝗮 𝘄𝗮𝘆 𝘁𝗼 𝘀𝘂𝗰𝗰𝗲𝗲𝗱? 🤔 Every software vendor wants customers to use theirs and no other software. Structural software suppliers are no exception. I've been doing this job for a long time and have seen many attempts to have all-in-one. How many do you think succeeded? Not one! I'm quite surprised that anyone is still trying to do this. It's a futile effort, usually leading to taking dangerous shortcuts and half-hearted solutions. There are many suppliers in our industry, each doing slightly different things and each good at something different. We can only get the best overall solution for structural engineers by interconnecting the best tools from more vendors. ➡️ 𝗦𝗵𝗮𝗿𝗶𝗻𝗴 𝗮𝗻𝗱 𝗰𝗼𝗹𝗹𝗮𝗯𝗼𝗿𝗮𝘁𝗶𝗼𝗻 𝗮𝗿𝗲 𝘁𝗵𝗲 𝘂𝗹𝘁𝗶𝗺𝗮𝘁𝗲 𝘃𝗮𝗹𝘂𝗲 𝘁𝗼𝗱𝗮𝘆. 𝗔𝗹𝗹-𝗶𝗻-𝗼𝗻𝗲 𝗱𝗿𝗲𝗮𝗺𝗲𝗿𝘀 𝘄𝗶𝗹𝗹 𝘀𝘁𝗮𝗻𝗱 𝗮𝘀𝗶𝗱𝗲. #structuralengineering #innovation #inspiration

54

5 Kommentarer

I get super annoyed every time I hear some clown say `const` was a mistake in JS and we should all just use `let`. It's annoys me because people listen. But this is a bad take. Writing code is more about communicating to the human reader than it is instructing the computer. Most of these takes about `const` at some point make the argument that `const`s in JS are not constant. By that they mean that any otherwise mutable value assigned to that constant can still be mutated after the assignment. And they are right! But just because `const` is functionally "immutable" in some languages, does not mean it means it is in all languages, and it does not mean that in JS. In JS, `const` means a constant _reference_, not a constant value. Now you may not like that. I would much prefer a constant value type than a constant reference type -- I eagerly await Records and Tuples for that -- but that does not mean a constant reference type is without value. More than that, `const` has taken on a broad and shared industry understanding that we _intend_ this to be treated like a constant value. The no-const-ers will show code like this: ``` const someArray = []; someArray.push('New array item'); ``` Then they'll argue something like: "Isn't that confusing? I wouldn't expect that to work, but it does! `const` isn't really a constant, so we shouldn't use it." They'll then suggest that `let` is better, because it doesn't lie about what it does (set aside that in some languages, let is immutable). But what's the alternative? I've seen arguments that you have no other option but to wrap your technically mutable data structure in Object.freeze, and you must do that recursively, otherwise, you should be using let. This is absurd. It's like these nay-sayers have never heard of a convention before. Before we had private methods in JS, we prefixed methods we _intended_ to be private with `_` (`_somePrivateMethod`). There were ways that existed that we could have used to *make* those methods private, but they were cumbersome and required enough effort and indirection that readability suffered, so we all agreed to abide by the `_` prefix convention. And it worked! Would we have preferred to not have to do that? Sure. Were we all excited when we got proper private methods? Sure! But in the mean time it was a good compromise. No one was out there saying "Don't use the `_` prefix convention because it's a lie! Instead every method should be thought of and treated as public because that's what it is!" That's what is being argued about `const` though! `const` isn't perfect, by any stretch, but it's a widely understood convention that communicates to other human readers our intent and it gets the job done right now. It's certainly a hell of a lot better than not communicating that intent at all by using `let` or sacrificing readability with recursive Object.freeze everywhere. I will rejoice when Records and Tuples arrive. In the mean time, I will be using `const` heavily.

15

6 Kommentarer

Sometimes when debugging, what you need isn’t another breakpoint in your firmware. Sometimes, what you need is just a little break from whatever you’re currently doing. In fact “Give up now, and try again later” can be surprisingly good advice. There are many forms of break: sleep, away from keyboard, eating well, doing chores, talking to your friends/coworkers, even talking to a duck (yes, it’s real. It’s called Rubber Duck Debugging method, Google it). If you’re too close to the problem you’re trying to solve, you can lose sight of the big picture. Your brain may get locked into analyzing only what's in front of you, missing the entire picture of your firmware. Giving yourself a break will break that vicious cycle. Just give up, and try again tomorrow, maybe the dream in your sleep will give you some hints, just like what I experienced. What? You’ve never experienced that? It’s like when you sleep, you dream about your problem, and suddenly you know where to look, and what to do. Upon waking up, you apply that solution and it’s just plainly working, flawlessly. Little did you know, when you sleep, your brain doesn’t stop processing your thoughts. Your brain enters creative mode (instead of pure analytical mode) to solve the issue. Sometimes creativity can solve the issue. It’s an effortless experience as if you have your brain think for you without you ever having to think. Repost if find this useful ♻️

106

7 Kommentarer

🔍 𝗨𝗻𝗱𝗲𝗿𝘀𝘁𝗮𝗻𝗱𝗶𝗻𝗴 𝗮𝘀𝘀𝗲𝗿𝘁() 𝘃𝘀 𝗿𝗲𝗾𝘂𝗶𝗿𝗲() 𝗶𝗻 𝗦𝗼𝗹𝗶𝗱𝗶𝘁𝘆 🤔 In Solidity, both assert() and require() are used to handle errors, but they serve different purposes! Knowing when to use each can make your smart contracts more secure and efficient. 🔒 𝗿𝗲𝗾𝘂𝗶𝗿𝗲() – 𝗩𝗮𝗹𝗶𝗱𝗮𝘁𝗶𝗻𝗴 𝗘𝘅𝘁𝗲𝗿𝗻𝗮𝗹 𝗖𝗼𝗻𝗱𝗶𝘁𝗶𝗼𝗻𝘀 Used to check conditions before execution (like user input, access control, etc.). Reverts the transaction if the condition is false and refunds gas. Example: Verifying that the caller is authorized or inputs are valid. require(msg.sender == owner, "Only owner can withdraw"); ⚠️ 𝗮𝘀𝘀𝗲𝗿𝘁() – 𝗖𝗵𝗲𝗰𝗸𝗶𝗻𝗴 𝗜𝗻𝘁𝗲𝗿𝗻𝗮𝗹 𝗜𝗻𝘃𝗮𝗿𝗶𝗮𝗻𝘁𝘀 Used to check internal conditions that should always be true (like contract state consistency). If the condition fails, it indicates a bug and halts execution, consuming all gas. Example: Ensuring balance consistency after a withdrawal. assert(balance[msg.sender] == previousBalance - amount); 𝗪𝗵𝗲𝗻 𝘁𝗼 𝘂𝘀𝗲 𝘄𝗵𝗶𝗰𝗵? Use require() to validate inputs or external conditions before proceeding. Use assert() to ensure the integrity of the contract’s internal logic. By understanding these differences, you can build safer and more efficient smart contracts! 🚀 #Solidity #SmartContracts #Blockchain #Web3 #Crypto #Ethereum #DeveloperTips

12

1 Kommentar

Last week I made a quick note about the setup of OneDev, an Open-Source suite of DevOps tooling very similar to Azure DevOps/TFS.  With as many features as it has, especially in the free Community tier, I turned that it into a full three-part series.  The first, mentioned last time, "OneDev: Part 1 - Setup and Usage" covers a basic setup and usage and product tour. I then followed it with "OneDev: Part 2 - Artifacts Agents and Docker" where I (try then mitigate) setting up a proper HA Database backend, CICD Pipelines and look at what can be done with artifact management. Wherever I get blocked, I do show work-arounds using things like Harbor CR and Pull-based replication of Git Repos with Forgejo. Lastly, in "OneDev: Part 3 - Advanced settings and features" I cover a whole lot of additional settings and features like secrets management (field configurations), Issue Configuration Templates and Email Templates.  I cover Discord integration with Notifications as well as User Impersonation.  I touch on some maintenance and tried to get the paid feature of Time Management to work. I think the TL/DR is that OneDev is a pretty solid offering and the community edition which has most features is free without limits.  The paid option is NOT a SaaS, rather a self-hosted with extra features and email support (and runs $6/User/Month bought in 12-mo increments - so you can pay for 12 users for 1 month or 1 user for 12 months - it works out the same). - "OneDev: Part 1 - Setup and Usage": https://lnkd.in/g2g64ZnG "OneDev: Part 2 - Artifacts Agents and Docker": https://lnkd.in/g_PEmrqb - "OneDev: Part 3 - Advanced settings and features": https://lnkd.in/gk64f9Bu - OneDev: https://onedev.io/ #OneDev #CICD #OpenSource #Kubernetes #Containers #AzureDevops

5

When evaluating engineers: Soft skills > Coding skills. In software engineering, we often get caught up in the technicalities - the languages, the algorithms, the data structures. But is that all there is to it? Let's take a step back and look at the bigger picture. What is the ultimate goal of software engineering? It's about creating value for customers, using technology as a tool. When hiring software engineers, it's easy to get swayed by impressive tech skills. But I urge you to first look for a product mindset. This is someone who understands that their role is not just to write code but to solve problems and create value. Next, look for creativity. In a field that's constantly evolving, the ability to think outside the box is invaluable. Enthusiasm is another critical trait. A passionate engineer will not only do their job but also do it with enthusiasm, constantly striving to improve and innovate and spread the energy to the rest of the team. Tech skills? Yes, they're important. But they can be learned and honed over time. Remember, we're not just building software. We're building solutions. We're building value. And to do that, we need more than just coders. We need problem solvers, innovators, and passionate creators. So, the next time you hire a software engineer, look beyond the code. Look for someone who can bring value to your product and customers. What do you think? I'd love to hear your thoughts and experiences. #softwareengineering #hiring

53

3 Kommentarer