Joey F. Poon

🤖 AI x PHP - Neuronv2 🤖 The release of Neuron V2 framework enables building real multi-agent systems in PHP - using event-driven workflows, streaming updates and human-in‐the‐loop interventions. "This implementation demonstrates that sophisticated AI architectures are no longer exclusive to Python. For teams with substantial PHP expertise, Neuron v2 removes the primary arguments for switching to Python for AI development. Complex agentic workflows are now feasible within the PHP ecosystem, using familiar patterns and deployment infrastructure." Has PHP has finally become viable for serious agentic / AI architectures, not just Python? https://lnkd.in/eWpFerVq

12

The difference between a “Senior Engineer” and a real Staff-level builder in 2026: A Senior Engineer ships tickets. A Staff Engineer ships outcomes. Here’s what I’m seeing in the market right now: • Seniors can execute within a defined roadmap • Staff engineers challenge the roadmap • Seniors optimize code • Staff engineers optimize systems + teams • Seniors build features • Staff engineers decide what not to build In AI, FinTech, and SaaS especially, founders aren’t hiring for “more hands.” They’re hiring for: Architecture ownership Production judgment 0→1 ambiguity tolerance The ability to reduce risk before it becomes visible The market is quietly separating engineers who can operate without guardrails from those who need them. And that gap is widening. If you’re hiring at Staff / Principal level this quarter, you’re not looking for 100% keyword overlap. You’re looking for signal: Can this person carry weight when it matters? Curious what others are seeing at the senior end of the market.

9

I asked Philip Su, early engineer at Meta and OpenAI, what he thinks of “vibe coding” and how we should think about the concept. “It’s easy to vibe code something to X. The hard part is setting up your App Store account… signing certs… approvals. So much of software is actually the business of software.” "But the other part in fairness, is that for many things in software, you don't need to ship the product. AI has made it so that you can vibe code yourself little tools all day....You can imagine software being much more broadly distributed because the everyday person can now create little tiny tools for themselves." "And I think that future is amazing! And the vibe-coding doesn't have to be 100% correct for that to be valuable." Philip, now building startup Superphonic breaks down vibe coding perfectly. Full episode drops tomorrow 🔥

8

1 Comment

The Truth About $100k Software Engineering Jobs In Japan... Yes, they exist — but not in the way most people imagine. I just posted a reaction video to a breakdown by Kei Fujikawa, a Tokyo-based software engineer with 10+ years in the industry, who explains who actually earns $100k+ (≈¥15m) in Japan. (Check in the comment section) A few uncomfortable realities: • The average dev salary (~¥5–6m) is real — and it includes many experienced engineers • $100k in Tokyo sounds huge, but after tax, rent, and lifestyle creep, it’s far from “easy mode” • Traditional Japanese companies still pay by age, not output • The real money is concentrated in a small set of companies: – Global tech (Google, Amazon, Microsoft) – Japanese mega-ventures (SmartNews, PayPay, Woven, etc.) – Highly specialised roles (AI, ML, bilingual EMs) One point that stands out: For senior roles, English + experience can matter more than Japanese in certain environments. That doesn’t mean Japanese isn’t valuable — it means the market is segmented. This isn’t about manifesting a ¥15m salary. It’s about understanding where you fit in the Japan tech market and choosing your path accordingly. If you’re serious about working in Japan, I recommend following Kei Fujikawa. He’s sharing grounded, insider perspectives — not fantasy takes. Curious: • If you work in Japan, does this match what you’ve seen? • For foreign engineers — has English helped you move up?

133

15 Comments

Game changer: Claude Code paired with Playwright CLI Playwright allows computers to work with the visual hierarchy of web pages, interact with page elements and take screenshots. Other ways to interact with web pages from Claude include the Claude Chrome extension and the Playwright MCP server. Both options are heavy on token use because they force the LLM to process images and a large MCP surface area. With the Playwright CLI you get token efficient browser processing. By default it's headless but you can also have it show what is seen in a window. Some ways Claude + Playwright CLI helped me recently. Testing Some AI generated code was completely broken on some API calls. The problem was the data shape was wrong so the tests "passing" meant nothing. Adding Playwright in the loop provided Claude a way to test the code in the same way a user would use the app. Iterative Design I pointed Claude at the current state and the reference for a "good" design I wanted to move towards. I interacted in real time regarding how the page was looking and directed test cases to further check behavior. I am currently exploring other use cases and I'm excited for a functional level up in my workflow. Are you doing visual work with AI? What tools are you using?

5

React still rules - but for how much longer? I'm seeing a real shift in frontend strategy lately 👀 More teams are adopting #Svelte and #Astrojs - not just as experiments, but as serious alternatives to #React, #Angular, and #Vue 📈 Why? ⚡ Faster performance 🧠 Simpler developer experience 📦 Smaller bundles, leaner code It’s not just a tech choice - it’s impacting hiring, delivery, and velocity 🤩 If you're building high-performing frontends and feel like the old frameworks are slowing you down, you're not alone 🤗 Curious to hear from others seeing (or driving) this shift - and if you need engineers already working with this stack 🤔 I can help with that 🫡 #techrecruitment #amsterdamjobs #frontenddevelopers

14

1 Comment

Everyone knows Firebase. Most of us have used it for years. But have you ever tried Supabase? 🤔 Honestly, this is not just about being “open-source” or “a Firebase alternative”. The story is much bigger than that. Personally, I needed a proper PostgreSQL database (and for free), and I needed a real-time system that actually works without draining my credit card on every single read. And that’s where things got interesting. The price difference is insane, and the flexibility you get with Postgres really makes you rethink your architecture on every new project. So I wrote two articles explaining the whole story: ✅ Medium (Detailed Deep Dive): For those who want the technical details, real cost comparisons with numbers, and how real-time actually works under the hood. https://lnkd.in/dwqRJ8RE ✅ Dev.to (TL;DR version): Straight to the point, no headache, no overkill details — when to choose this and when to choose that. https://lnkd.in/d_pzkm-c If you’re tired of bill shocks and want real control over your data and costs, this is an experience worth reading. #Supabase #Firebase #WebDev #PostgreSQL #Backend #Programming #TechTalk

4

Marco Roth's Herb ecosystem released version 0.8 last month, bringing sophisticated tooling specifically for HTML+ERB files - the foundation of most Rails view layers. Major improvements: 🔹 Rust and Java bindings expanding Herb beyond Ruby ecosystem 🔹 Configuration file support (.herb.yml) for team-wide linting and formatting 🔹 Autofix capabilities for linter offenses 🔹 Tailwind CSS class sorting integration 🔹 Language Server improvements with fix-on-save and code actions Why this matters for Rails careers: Developers working with modern Rails tooling demonstrate awareness of ecosystem evolution. Understanding Herb's capabilities - particularly linting, formatting, and Language Server features - signals professional approach to view layer development. The move toward better HTML+ERB tooling reflects Rails' commitment to server-side rendering whilst maintaining developer experience comparable to modern JavaScript frameworks. Integration benefits: 🔹 Visual Studio Code extension with full LSP support 🔹 Rails Dev Container integration by default 🔹 Local CI capabilities for faster test feedback 🔹 Performance monitoring through deep HTML+ERB understanding Companies building ambitious server-side rendered Rails applications increasingly value developers familiar with modern view layer tooling. For Rails developers focused on traditional server-rendered applications, Herb represents significant quality-of-life improvements without requiring JavaScript framework adoption. What view layer tooling has most improved your Rails development workflow? We're connecting developers who understand Rails' modern server-rendering capabilities. #Herb #RubyOnRails #ViewLayer #DeveloperTools #TechRecruitment

3

KOMOJU has told us they're rapidly scaling up their customer engineering team. They're open to candidates of all technical skill levels. Japanese ability is required, but being able to communicate using the language is more important than speaking it perfectly. For instance, they currently have a non-Japanese engineer leading customer facing communication, while a native Japanese one handles internal communication. This was impressive to me, as I'd normally expect the opposite. We interviewed their head of engineering recently, which gives a good overview over what the role entails. https://lnkd.in/gKi6m9rV

60

2 Comments

We just released Taro's newest course about what life is really like at FAANG. So much of the discussion about working at FAANG focuses on things like the scale of the work or the free snacks. But on the day-to-day level, an engineer at Google or Facebook doesn't think about these topics. They are spending most of their waking hours on things like: - how much scope do they have? - how much control do they have vs just being a cog in the machine - what do their coworkers think about them? As a 5 year veteran of Meta, I break down the real pros and cons of working in Big Tech. Check it out here: https://lnkd.in/gfkwVDf9

58

4 Comments

Hiring senior engineers at early stage is one of the biggest challenges we face. You need people who can go from zero to one, wear five hats, and still ship clean code. Meanwhile, your company isn’t a household name yet, and you’re competing with FAANG comp packages. That’s where founders burn out — spending nights sourcing, chasing intros, interviewing, only to have a candidate decline your offer. Allen Osgood (founder of Eisen) found that listening to candidates' needs and building a plan to support them worked best! “The biggest thing I came to appreciate during closing was finding out what’s important to this person… Some candidates have family situations that require higher cash, lower equity. Others really care about growth, learning budgets, opportunity. It’s about tailoring your entire closing process to their incentive set, life situation, and preferences.”

27

1 Comment

For over a decade, React has lived in a single-threaded universe: one JavaScript thread handling events, logic, and rendering. But what if React had been designed differently. What if it embraced two threads from the very beginning? At React Universe Conf 2025, Xuan Huang, Ziqi Zhen, and Haoyang Wang from ByteDance introduce React for two threads, a paradigm shift in how React applications can be built and executed. ▶️ https://lnkd.in/dUKWdb96

5

Paraphrased from a job ad I just read: "This role requires engineers who are intimately working with Claude Code or Cursor, delegating entire implementation tasks to agents, and using Claude skills or project context to maintain quality with increased speed." The company that posted isn't even a particularly innovative company, they hire 'decent' engineers at best. They're also a larger, more process heavy company than many smaller outfits. And yet, there are genuinely innovative startups and smaller scaleups in Australia who (as I've heard along the grapevine) are massively struggling to see any positive impact from AI in their engineering teams whatsoever. Some are trying to push their engineers to adapt, but are failing. Others are simply ignorant, I heard of one in particular who suggested "our tech is too complex for AI to help us" - if you knew who I'm talking about, you'd agree that their business is about as complicated as a paperclip... ^this is an ego issue, of course. Anyway, these are the businesses who should be in pole position to take advantage of what's happening right now. They're small, flexible, supposedly filled with the best talent in the country, and yet they're stagnating or slipping backward whilst so many others level up. There is going to be a major reshuffling of the cards in our ecosystem over the coming months to year or two... Make sure you're working for a company that's doing everything necessary to end up at the top of the deck.

22

17 Comments

Most people use Claude Code wrong. 🤖 They treat it like ChatGPT with a terminal. But the real power is in context control. If you’re not configuring your CLAUDE.md properly, you’re wasting tokens 💸 and getting generic output. Your CLAUDE.md should clearly define: • 🏗 Project architecture • 🛠 Tech stack • 📁 Folder structure • 📐 Coding standards • 🏷 Naming conventions • 📜 “How we build here” rules Think of it as onboarding Claude like a senior developer joining your team 👨‍💻 The better the context, the smarter the output. AI coding tools aren’t magic ✨ They’re leverage. And leverage depends on configuration. How are you structuring your CLAUDE.md? 👇 #AI #Claude #DeveloperTools #BuildInPublic #StartupLife #SaaS #TechFounder #Productivity #Engineering

10

1 Comment

You can backfill a Senior Engineer in 6 weeks. A great Engineering Manager? That search will humble you. I've watched brilliant CTOs make the same mistake. They treat the EM search like a senior IC search: post the role, screen for technical chops, pick the sharpest coder who also "seems like a leader." Six months later, the team is in shambles. Here's the uncomfortable truth: Engineering Managers are the load-bearing wall of your entire org. They're the ones absorbing pressure from the top, protecting velocity at the bottom, and keeping your best engineers from quietly updating their LinkedIn profiles. Get it wrong, and you don't just have an open seat. You have a retention crisis, a morale problem, and a roadmap that's bleeding out. The CTOs I respect most have figured something out — the technical bar for an EM matters far less than their ability to build psychological safety, navigate ambiguity, and make their team feel seen. That combination doesn't show up on a resume. It shows up in a conversation. If you're in the middle of an EM search right now or about to kick one off, I'd love to share what we're seeing in the market. What's working, what's not, and where the real talent is sitting. Because the best Engineering Managers aren't on the job boards. They're running standups right now for someone else. What's the one quality that separates a good EM from a great one in your experience?

20

1 Comment

**Agentic workflows** are coming to every engineering org. The difference between “wow, huge leverage” and “what did we just break?” won’t be the model. It’ll be **operating discipline**. In plain terms: you’re giving software the ability to take multi-step actions—open PRs, modify files, trigger CI, run tasks. That’s powerful… and operationally expensive if unmanaged. Where teams usually slip: - Giving agents **developer-admin** privileges - Treating output like “suggestions” instead of **change events** that need controls Operator rule: treat agents like **new employees who never sleep**.   They need onboarding, permissions, supervision, and an audit trail. A rollout playbook that works: - Start with **one repo / one workflow** - Define **allowed actions** explicitly - Require **human review** for merges + deploys - Log every tool invocation + artifact Guardrails to implement early: - **Least-privilege** tokens (scoped + time-limited) - Clear **secrets boundaries** - **Budgets + rate limits** (cost ceilings) - **Environment separation** (dev/test vs prod) If you want agents to ship value: don’t “install and pray.”   Design the workflow like you’d design payments or production access: controls, visibility, fallback. Source: https://lnkd.in/g3QYtu7s

1

We’re continuing 30 𝐌𝐢𝐧𝐮𝐭𝐞𝐬 𝐭𝐨 𝐌𝐚𝐬𝐭𝐞𝐫 𝐑𝐞𝐚𝐥-𝐓𝐢𝐦𝐞 𝐏𝐢𝐩𝐞𝐥𝐢𝐧𝐞𝐬 𝐟𝐫𝐨𝐦 𝐒𝐜𝐫𝐚𝐭𝐜𝐡 with Shih-min(Stan) Hsu in 2026! Three new blogs to deepen your understanding of stream processing: • Database Lookup Join for Stream Processing: Architecture, Implementation and Limitations: https://lnkd.in/gvHX_y2a • From Lambda to Kappa: The Evolution of Stream Processing Systems: https://lnkd.in/gwXkMBCT • Giving Stream Processing a “Memory”: The Power of Stateful Operations: https://lnkd.in/g5yfDBca Happy reading! #risingwave #streamprocessing #realtimedata

10

𝐕-𝐓𝐞𝐜𝐡𝐮 𝐇𝐮𝐛 #𝟏𝟕: 𝐂𝐡𝐫𝐨𝐦𝐞 𝐃𝐞𝐯𝐓𝐨𝐨𝐥𝐬 𝐌𝐚𝐝𝐞 𝐄𝐚𝐬𝐲: 𝐃𝐞𝐛𝐮𝐠𝐠𝐢𝐧𝐠 𝐖𝐢𝐭𝐡𝐨𝐮𝐭 𝐭𝐡𝐞 𝐇𝐞𝐚𝐝𝐚𝐜𝐡𝐞 Your local build feels fast, metrics look perfect - yet performance issues keep showing up on real devices. The truth is, localhost doesn’t reflect real user experience. In the December issue of V-Techhub, Minh Nguyen – Senior FE Software Engineer, dives into Chrome DevTools’ Performance panel, breaks down Core Web Vitals (LCP, CLS, INP), and puts the AI Assistant to the test to see if it actually speeds up debugging. 👉 If you’re building for real users, this one’s for you: https://shorturl.at/Nu7a2 ---------- 𝐕𝐢𝐧𝐨𝐯𝐚 - 𝐏𝐚𝐬𝐬𝐢𝐨𝐧 𝐟𝐨𝐫 𝐏𝐞𝐫𝐟𝐞𝐜𝐭𝐢𝐨𝐧 Award-winning IT consulting and development powerhouse in Singapore More about us: www.vinova.sg

1

What a “10x Software Engineer” Really Is We’ve all heard the phrase “10x engineer.” It’s usually misunderstood. A 10x engineer isn’t someone who types faster, ships more lines of code, or works unsustainable hours. The real 10x multiplier comes from something far more rare—and far more valuable. 1. They create leverage, not code volume A true 10x engineer builds systems, abstractions, and tools that make everyone around them more effective. Their impact compounds through the team. 2. They have ruthless clarity They see the essence of a problem quickly—what matters, what doesn’t, and what the simplest workable solution looks like. This clarity prevents months of wasted effort. 3. They manage complexity with taste Taste is the underappreciated trait of elite engineers. It’s the instinct for clean architecture, thoughtful tradeoffs, and code that’s built to evolve, not just to work today. 4. They raise the bar by how they work Their communication is crisp. Their reviews teach, not nitpick. Their sense of ownership is contagious. They model a standard others naturally adopt. 5. They make the whole team move faster This is the real multiplier. The best engineers unlock speed across product, design, infra, and leadership by reducing uncertainty and increasing confidence in decisions. A 10x engineer isn’t a solo superhero—they’re a force amplifier. They don’t outwork the team; they elevate it.

8