Jerome Raymond

I used Claude in the terminal to fix Claude's own VS Code plugin setup. That moment felt like something had shifted.                                I'd been hearing the hype for months. Used ChatGPT, Copilot, GitLab Duo. Good tools. But nothing that made me think "this is fundamentally different."                                     Then I spent 4-5 days actually using Claude Code. Here's what's different — it doesn't wait for you to attach files or spell out context. It figures out which files are relevant, asks permission to read them, and gets to work. Less hand-holding. More like working with someone who actually looks around before asking questions. The planning mode changed how I approach implementation. Before writing a single line, it drafts a plan you can push back on, comment on, reshape. The plan updates. Then it builds. That's not just speed — it's confidence in what you're building. And then there's the agent stuff. I'm now using a Jira MCP that lets Claude update my tickets based on what it's actually doing in the codebase. Not summarizing — acting. 4-5 days in, I'm still setting up MCPs, exploring hooks, figuring out what "power user" even means here. But it already feels less like a tool and more like an intern who doesn't just draft things — but takes actions, with your permission. I didn't expect to feel that so quickly. Are others on this journey? What's the thing that made it click for you? #AITools #SoftwareEngineering #ClaudeCode #DeveloperProductivity #AIAtWork

6

I’ve seen production keys leaked because someone hardcoded a value. Not a bug. A process failure. Good engineers write features. Great engineers protect systems. Environment variables separate: → Code from config → Secrets from repos → Environments from chaos” Environment variables aren’t optional. They’re foundational architecture. #NextJS #SecurityFirst #WebDevelopment #EngineeringMindset #BuildInPublic #DeveloperLife

10

Here’s the uncomfortable truth about why junior developers can’t get hired right now. Many juniors wonder why they can’t land their first role. The reality? You’re often competing with candidates who have 10+ years of experience and advanced degrees. Employers are naturally drawn to these “safer” bets—but that doesn’t mean juniors are out of luck. Understanding this gap helps you strategize smarter, not blame yourself. My take: focus on building projects that solve problems that would mimic real problems you would likely face in the workplace. Connect with senior developers, recruiters and even directors and start a conversation. This will help you figure out what “real problems” would look like and from there you’d be able to see what projects may help you land that next role. Question: Did you have a project that helped you stand out or land a role? #softwaredevelopment #softwareengineering #jobmarket #competition #tech

9

2 Comments

Tech friends, is anyone here actually using agent swarms for development? I mean automated, parallel agents building features, deploying them and testing them, end to end. If so, how. This *has* to be the next stage of software engineering but I can't find anyone doing it for real.

3

2 Comments

Walmart 𝗿𝗲𝗰𝗲𝗻𝘁𝗹𝘆 𝗼𝗳𝗳𝗲𝗿𝗲𝗱 𝗺𝘆 𝗼𝗻𝗲 𝗳𝗿𝗶𝗲𝗻𝗱 𝗮 𝗽𝗮𝗰𝗸𝗮𝗴𝗲 𝗼𝗳 𝟮𝟮 𝗟𝗣𝗔 Position: Frontend Engineer Application Method: Referral 𝗥𝗼𝘂𝗻𝗱 𝟭 - 𝗧𝗲𝗰𝗵𝗻𝗶𝗰𝗮𝗹 1️⃣ How would you design a highly performant web app that renders large data tables with real-time updates? 2️⃣ Explain how you’d reduce the initial page load time of a React application by 40%. 3️⃣ You notice a memory leak in a React app — how would you debug and fix it? 4️⃣ What’s your approach to ensure consistent rendering and layout across different browsers and devices? 5️⃣ How would you structure and optimize a large-scale React codebase to maintain scalability and readability? 𝗥𝗼𝘂𝗻𝗱 𝟮 - 𝗗𝗲𝗲𝗽 𝗧𝗲𝗰𝗵𝗻𝗶𝗰𝗮𝗹 1️⃣ How does React’s reconciliation algorithm (Fiber) work internally, and how does it improve rendering performance? 2️⃣ How would you implement dynamic theming (light/dark mode) in a large web application without performance issues? 3️⃣ A complex dashboard is lagging when users apply filters — how do you profile and optimize it? 4️⃣ How would you design a system to handle client-side caching, API retries, and error boundaries gracefully? 5️⃣ Explain your strategy for managing global state efficiently in a React app with multiple teams contributing to it. 6️⃣ How would you ensure accessibility (a11y) and WCAG compliance in your frontend application? 𝗥𝗼𝘂𝗻𝗱 𝟯 - 𝗠𝗮𝗻𝗮𝗴𝗲𝗿𝗶𝗮𝗹 / 𝗕𝗲𝗵𝗮𝘃𝗶𝗼𝗿𝗮𝗹 1️⃣ Tell me about a time you delivered a frontend feature under tight deadlines — how did you prioritize and execute? 2️⃣ How do you handle a disagreement with a designer or backend developer on implementation details? 3️⃣ Describe a time when your production build caused a major UI issue — what was your response and fix? 4️⃣ How do you explain frontend performance metrics (like TTI, FCP, CLS) to non-technical stakeholders? 𝗜 𝗵𝗮𝘃𝗲 𝗽𝗿𝗲𝗽𝗮𝗿𝗲𝗱 𝗖𝗼𝗺𝗽𝗹𝗲𝘁𝗲 𝗜𝗻𝘁𝗲𝗿𝘃𝗶𝗲𝘄 𝗣𝗿𝗲𝗽𝗮𝗿𝗮𝘁𝗶𝗼𝗻 𝗚𝘂𝗶𝗱𝗲 𝗳𝗼𝗿 𝗙𝗿𝗼𝗻𝘁𝗲𝗻𝗱 𝗘𝗻𝗴𝗶𝗻𝗻𝗲𝗿𝘀. 𝗚𝗲𝘁 𝘁𝗵𝗲 𝗚𝘂𝗶𝗱𝗲 𝗵𝗲𝗿𝗲 - https://lnkd.in/d2w4VmVT 💙- If you've read so far, do LIKE and RESHARE the post (👍)

412

14 Comments

I’ve worked with many remote teams over the years, and one pattern always stands out. The developers who get rehired the fastest aren’t the smartest — they’re the most reliable. They don’t just deliver code. They deliver confidence. They show up when they say they will. They communicate clearly when something’s delayed. They make clients feel like their project is safe. When you’re remote, trust is your currency. Every missed message or late reply chips away at it. Every clear update builds it back. In a world where clients never meet you face-to-face, your professionalism is your reputation. If you want long-term clients, don’t just write great code. Make people feel like working with you is the easiest decision they made that week.

77

22 Comments

Lately I’ve been thinking about how much of software development is just dealing with decisions made in the past. Sometimes your own, sometimes someone else’s. Working with .NET, debugging things, revisiting old code… it makes you realize pretty quickly that small choices don’t stay small for long. Still figuring things out, but trying to leave things a little cleaner than I found them. #dotnet #softwareengineering

35

2 Comments

Software engineers fear the AI that will take their job while the real danger I see should be in companies and solution owners themselves. Products that you spent years and tons of dollars for buildng selling it for whatever the number of subscriptions is can now be easily cloned and it costs nothing but the Gen AI subscription and some small resources. I am really curious of how the scene will look like after maybe Q2 of 2026. The development that business owners needs to follow “not to fail in the common nokia example” is now talking days not years like all used before!! #ArtificialIntelligence #GenerativeAI #SaaS #StartupLife #ProductManagement  #SoftwareEngineering #TechTrends #DigitalTransformation  #Innovation #FutureOfWork #Entrepreneurship  

5

2 Comments

In 12 months, coding agents went from writing none of Clipboard's code to nearly all of it. That broke our tests. At one point, 100% of PRs in two of our largest repos hit at least one flaky test. When humans write code, flakes are annoying. When agents write code, flakes break the feedback loop that keeps them moving at full speed. We drove our E2E flake rate from 100% to under 15% in six weeks: 1. We asked agents to triage every E2E test. Three models with separate harnesses categorized each, then two more agents reached consensus in fresh context windows. They proposed cutting 174 tests to 46. We landed at 87 after domain owners pushed back on specific cuts. 2. We built a Playwright reporter designed for agents with a unified timeline of steps/network/console events, base64 screenshots, and traceparent headers that let agents jump from a failed test straight to Datadog APM traces across 30+ backend services. 3. Agent selection matters. Given identical flakes and prompts, Codex consistently went deeper than the alternatives, returning trace evidence and real product bugs instead of defaulting to retries and longer timeouts. Code is a liability. Tests usually get a pass because "coverage is good." Except that each test has a maintenance cost, and you pay the highest cost for lying tests. Full write-up, plus our open-source @clipboard-health/playwright-reporter-llm and /flaky-test-debugger skill: https://lnkd.in/gMuJjbkk

85

9 Comments

Agent coding tools increase the blast radius of your developers. For your best developers this is a positive. But for your low performers? This is a negative development. The impact of your hiring has gotten even higher than it was before. A poor performer will not magically be a person who can get good results from these tools. But they will be able to churn out really poor quality solutions much faster than they could before. If your high performers can't keep up and you don't have good enough guardrails in place this is going to create real problems for your team and the company.

8

7 Comments

As developers, we often chase the big, shiny new features, but sometimes it's the subtle enhancements that truly streamline our daily grind. I've been diving into C# 14 and .NET 10 recently, and I'm excited about a few things that promise to make our code cleaner, more robust, and frankly, more enjoyable to write. These aren't just theoretical improvements; they're practical tools that can immediately impact our productivity and the quality of our software. It's about writing less code to do more, and doing it more safely. Take a look at this document over here. #CSharp #DotNet #DotNet10 #CSharp14 #SoftwareDevelopment #CleanCode #DevOps #Kubernetes #AKS #CloudNative #ProgrammingTips

29

2 Comments

🔥 STOP applying to .NET jobs the old way. Because resumes don’t get interviews anymore. Visibility does. If you’re a .NET developer struggling with callbacks, this will change your approach 👇 🧠 (Save this — you’ll reuse it) 💼 The 2026-ready .NET Job Strategy (that recruiters actually notice) 1️⃣ Pick ONE core identity Not “Full-stack dev”. 👉 .NET Core | Web API | Microservices | Azure Clarity > complexity. 2️⃣ Post what you’re learning (not what you know) Recruiters follow learning signals, not certificates. Examples: “Why I chose Minimal APIs over Controllers” “What broke when I scaled my .NET API” “EF Core mistake I won’t repeat again” 3️⃣ Comment smarter than you post Your comments are mini-resumes. Add: Insight Experience Trade-offs Not: “Great post 👍” 4️⃣ Build proof in public One small repo beats 10 certificates. Ideas: Auth API (JWT + Refresh Tokens) Clean Architecture sample Performance comparison (Dapper vs EF Core) 5️⃣ Be consistent, not loud 3–4 posts/week 5–10 meaningful comments/day That’s how the algorithm learns who to show you to. ❓ Question for you: What’s the ONE .NET skill you’re doubling down on in 2026? Comment below 👇 (I read every reply) 🔁 Repost to help another .NET dev ❤️ Like if this was useful ➕ Follow for daily .NET, system design & job-search insights https://lnkd.in/gqPbUNue #DotNet #DotNetDeveloper #CSharp #ASPNetCore #BackendDeveloper #SoftwareJobs #TechCareers #LinkedInAlgorithm

3

How to write the best prompts for a coding AGENT: Pair with another HUMAN on understanding and describing the problem. Still the same as before.

6

A quick toolkit for not losing exceptional software developers during your hiring process. I spend most of my time speaking with engineers about new opportunities. One thing that still stands out, over 60% of developers say they lose interest in a role because the hiring process is slow or unclear. Not the company. Not the tech. The process. Most of the time it’s small things that cause it. I’ve put together a quick summary awith 5 mistakes I still see far too often. Worth a looking over if you're hiring.

9

Over the past year, I’ve spoken to a lot of experienced developers who are exhausted with the modern job hunt. Not because they can’t code. Because the entire process increasingly feels disconnected from actual engineering ability. Senior developers are being ghosted after decades in the industry. Resumes are turning into keyword games. Recruiters are overwhelmed. Job boards feel noisy, repetitive, and impersonal. That frustration is a big part of why I started building SageLinked. The idea is simple: Create a platform that helps experienced developers present real capability, real experience, and real technical signal - not just ATS-optimized buzzwords. I’m especially interested in hearing from senior developers right now: • What frustrates you most about current job boards? • What functionality do you wish existed? • What would actually make the process feel fairer, smarter, or more human? • What makes you immediately distrust a recruiting platform? I’d genuinely love the feedback while the platform is still early enough to shape correctly. SageLinked page: https://lnkd.in/dbyrVm6W

1

🚨 IF YOU’RE APPLYING FOR A .NET ROLE AND CAN’T ANSWER THESE… YOU’RE NOT READY YET. Not trying to scare you. Just telling you what interviewers silently judge you on. Here’s what separates average developers from hire-worthy engineers C# DEPTH (Not Just Syntax) → Do you REALLY understand async/await (how it works internally)? → Value type vs Reference type — what happens in memory? → IEnumerable vs IQueryable — WHEN does the database get hit? If you can’t explain the WHY, you don’t own it yet. ⚙️ ASP.NET CORE REALITY CHECK → What actually happens when a request enters the pipeline? → Middleware order = behavior change. Why? → Minimal APIs vs Controllers — trade-offs? This is where most 0–2 year developers struggle. 🗄️ ENTITY FRAMEWORK (The Silent Filter) → How does EF Core Change Tracking work? → Lazy vs Eager loading — performance impact? → When should you use Dapper instead of EF? Performance awareness = engineering maturity. 🔐 SECURITY (NON-NEGOTIABLE) → JWT flow: login → token → validation → authorization → Where should tokens be stored — and why? → How do you prevent SQL Injection, XSS, CSRF? If you don’t know this, you’re not production-ready. 🏗️ ARCHITECTURE THINKING (THIS GETS YOU HIRED) → Clean Architecture = separation of concerns → When would you apply CQRS? → Horizontal vs Vertical scaling? This is what turns interviews into discussions. Truth: Average candidates memorize answers. Top candidates understand systems. If you can explain these clearly without hesitation, you’re already ahead of 70% of applicants. Save this before your next interview Follow for practical .NET prep #DotNet #CSharp #ASPNetCore #BackendDevelopment #FullStackDeveloper #InterviewPreparation

304

8 Comments

This is precisely what I expected and have mentioned in a prior post... somewhere. I'm not sure why some are surprised by this. We know that AI companies have been losing money. We know this is not sustainable. We also have loads of other examples of the bait, hook, and raise prices tactic. (One only needs to look at their own streaming subscriptions and their ballooning bills, along with *ads* being added.) AI is not a magical replacement for people. Even if it were, do you really want to let another company dictate what your AI employee's salary is without you having any say? Do you want to always be guessing if your next quarter will be profitable, in spite of rapid growth and adoption of your product? Do you really want to be part of the pyramid scheme in which costs for services will inevitably trickle down and only those on the very top will win?

1

The most dangerous "smell" in a video codebase? The 2,000-line Component. I see it constantly: a React or Vue component that handles UI rendering, ABR logic, analytics, and error recovery all in one file. It’s a ticking time bomb. When you couple UI to media logic, your system breaks every time you want to "move a button." To scale, you need a Headless Player architecture.  The Blueprint The Headless Player: Stop treating the player as a UI widget. It’s a singleton service. The engine manages the buffer and network; it doesn't even know the DOM exists. The Throttled Event Bus: Video events (timeupdate) fire 60fps. Don't pipe these into Redux or React Context. Use a Pub/Sub bus throttled to ~250ms to keep the main thread smooth. The Adapter Pattern: Every browser (Safari vs. Chrome) is different. Use an Anti-Corruption Layer so your app says engine.play(), and the Adapter handles the browser-specific quirks. The Mandate: Isolate the volatile network logic from the volatile UI. Build for Change. Build a Headless Engine. #SoftwareArchitecture #SystemDesign #CleanCode #VideoEngineering #WebDev #videoPlayer #nodeJS

3