Maria Lourdes Indiano Järnefelt

Localization is more than translation. Pendo.io just rolled out our first localized experience—in Japanese 🇯🇵—designed for Japan’s fast-growing, AI-driven tech market. Go Pendo.io Japan ! From login screens to product areas, we’re bringing product-led growth to users in their language. Learn how we're doing it—and why Japan was first! 👇

2

This is a perspective I strongly agree with 👇 The idea of engineers spending time in a Forward Deployed Engineer (FDE) role is underrated. At its core, it forces you to move beyond just writing code → into understanding why the code exists in the first place. This shift is becoming even more important in the AI era. Because writing code is getting easier. Understanding the problem is not.

1

Documentation Hub is now live on Parallel AI! Training and documentation right inside the platform. Soon we will allow white-label owners to contribute and create their own documentation guides for the ecosystem and their specific platforms.

6

There’s nothing quiet about what’s happening at Omnisend 🙉 We expect everyone to be sharp with AI. Not just the engineers or the usual innovators. Everyone — from product to marketing to payroll. We don’t treat AI like magic. Or even a novelty, to be fair. We treat it like the smartest people treated the internet, or version control, or the printing press — a tool that’s here to stay. You either learn to use it, or you slow the team down. That’s why AI Day at Omnisend is still going. It’s evolving. Getting more serious. A reminder that being good with AI isn’t someone else’s job — it’s yours. By the fourth edition, the shift’s obvious. Fewer people asked what should they be doing with AI?, and instead, they got together and built things — tools that improve the product, improve their day-to-day, or just remove the boring parts of the job. And a lot of what they built is actually shippable (if not shipped already). The goal really isn’t full automation. The goal is automating the boring, letting the human creativity loose, and exploring the parts of our work that remind us why we even chose these professions. P.S. 5th edition of AI is coming up real soon, let's see what will that one bring.

35

1 kommentar

We’re living in a crazy time. Today, you can set a model running in a 24-hour feedback loop to implement a complex module with tests and documentation, work that used to take a week or more. That changes the game. Standing out is no longer about being the best coder in the room or juggling the most tickets. More and more, the real differentiator is whether you can think like an architect, act like a product manager, and lead like a teammate people trust.

14

1 kommentar

This week has been a long, tiring, but exciting week. The UA day event that took place online globally gave me much to think about and look forward to. But even more exciting was finally rolling out a #ClaudeSkill to my product team, which I was working on for the past few weeks. The aim of the skill is simple enough - to generate all kinds of UI texts that are required for a product feature - but the entire process of #vibecoding the skill felt fascinating! Steeping the skill with standards and guidelines, dos and don’ts, terminology, tone and style, and so many other aspects that I wanted it to consider - was indeed very interesting. Staring off into space yesterday, I found myself thinking of more ways to improve the skill! 😂 With preliminary feedback from my teammates who’ve already used the skill, the results look great! Having left the world of coding 13 years ago, it was amusing to “talk” about performance improvements and fix security issues with Claude Code. 🤣 The pace of “AI” is changing every minute, but I guess it’s up to us to find things to do that excite us even in this chaos. Akshay Avinash - Thanks for helping me with my idea from day one - from helping me install the damn thing - to always being available to answer my zillion questions - to testing every version of the skill in between - to verifying my roll out mail as well! 😬 It is thanks to people like you at #SAP that I enjoy what I do everyday! 🎉

21

I have been testing OpenCode Go’s new low-cost subscription, and it’s become a permanent part of my workflow, not as a replacement for my main tools but for a specific task of structured thinking work. This includes PRDs, user stories and clear specifications for coding agents. Kimi K2.5, one of the included open-weight models, excels at this. I also quickly tested it on a TypeScript Next.js project and achieved solid results with only minor errors. The full model lineup includes GLM-5, Kimi K2.5, MiniMax M2.5 and MiniMax M2.7. If you’re building with AI tools and need a low-friction option for the thinking layer and not just the coding, this subscription is worth it.👌https://opencode.ai/go

1

1 kommentar

Proud of the work from our team on enhancing the safety and maintainability of our frontend by moving away from React.FC. This shift is significant for modern React and TypeScript codebases and shares practical lessons on scaling UI systems while improving developer velocity. This is a great read for any frontend engineer focused on long-term code health. Read more here: https://lnkd.in/g8v6k8Gs

232

12 kommentarer

Just shipped our most requested feature at Skarbe. After months of development and hundreds of customer requests, we've launched our Zapier integration. Why is this a big deal? Because it unlocks thousands of automation possibilities for our users. They can now connect Skarbe to virtually any tool in their stack. But here's what really matters: Our customers can finally build the exact workflows they need. No compromises. No workarounds. Just seamless automation that works for their unique processes. Behind the scenes, we've also been steadily improving our core product: - Smoother navigation - More intuitive interfaces - Faster response times Small changes that add up to make a big difference. The truth is, building software isn't about big, flashy features. It's about listening to your users and consistently delivering value. Sometimes that means major releases like Zapier. Sometimes it's those tiny UX improvements nobody notices but everyone benefits from.

25

5 kommentarer

Teledyne SP Devices now builds and ships complete multi channel digitiser systems (PCIe or PXIe) with a wide range of GPU, motherboard, CPU and SSD options. You configure according to your digitiser and channel count requirements, and we will build, test and ship a complete multi channel industrial system in a rack or tower configuration. Systems with 4 to 128 channels already built, configured and shipped for various scientific and industrial applications Join the upcoming webinar to learn more.

2

Been using Temporal since 2020 at Twilio for handling ML system failures. Production ML is unreliable—constant crashes, timeouts, random errors. Temporal's Durable Execution made our code crash-proof. Instead of building custom retry logic and error handling, we could just focus on the actual application. I can't wait to start integrating the new SDK!

1

Weekly reminder that w/ AI people produce more value - grow - boost quality - don’t optimize for your current targets. Aimless production does not produce value; more than ever we have to be focused on the right production to win.

9

Working on DevEx these days does of course also mean enabling people to use AI. After Gustav Söderström mentioned Honk externally the other day, we have gotten a ton of questions - what is it and can I use it, can I also 🪿honk🪿? To no surprise Honk is an internal name for our tool (and a verb as well by now) where we combine fleetshift with AI coding agents. The goal has always been to be able to run complex code changes across the software fleet, to be able to complete these drawn out migrations we never were able to invest the time in to complete. Putting an AI coding agent into our fleet management system makes it possible to go one component at a time, let the agent make a code change and post that PR for review and deploy. A fun thing happened though - when we made it super easy to run background coding tasks we also started to see a shift in how we do things and what we do. In addition we see many small things being fixed that earlier wasn't always adressed, so not only are we completing these long running migrations, we also fix the small things to improve our code. And of course, also building features in the background, like coming up with an idea at the subway and having it implemented and done when arriving in the office, ready to be picked up and tried out on the phone. Max Charas and I put together this little webinar - I hope it answers the question what the 🪿 it is! https://lnkd.in/dSdkTwuk

96

1 kommentar

Do you know what "Spec Driven Development" is? In my humble opinion, it is nothing new, but it is becoming a new buzz word now that AI can spit out code faster than we can ever read it. The, not so novel, idea is to focus first on writing good quality specs for the products we are building and then having a workflow that goes all the way from product requirements to low level task lists so we delegate to AI agents. The two main products out there focusing on that are the OSS Github Spec Kit and a whole new IDE (another VSCodium based one though) from AWS called Kiro. Last week I got access to Kiro and I added it to the list of tools I'm reviewing on the "Vibe Code With Me" series and the "raw" long video is out if you are interested: https://lnkd.in/d5wW3Spn If you want to be notified when I'm done with this long videos and just want the smaller summarized video in the end, leave a comment here so I can notify you when that video is out (it will take a while, there are a lot of tools I want to check out first). #vibecoding #aws #kiro

9

In my 10+ years with Pimcore at Studio Emma, this is the biggest leap I've seen yet! Not only is the long-awaited Pimcore Studio UI being released, but they’re also introducing an Agent SDK. Pimcore is now being built for both humans and agents. This means agents can work with the full context of your data, seamlessly, whether integrated or external. We’re not looking into the future anymore, we’re standing right at the edge of it!

29

Last week in PyTorch Conference 2025, we announced we are open sourcing torchcomms and NCCLX/CTran. Today, we share more details about NCCLX/CTran design and how we used them in production GenAI training and inference. Check out our white paper on this topic - https://lnkd.in/gySuXi6Y Some features we covered in this paper - * Host-driven collectives * Zero-copy data transfer * CTran/Network co-design (DQPLB) * Zero-copy and SM-free Send/Receive for PP * RMA Put for TP * Fault tolerant AllReduce * GPU-resident collectives for EP * Low-latency optimization * Scalable Initialization in Training * GPU Memory Management for comms * Fault localization and Performance Observability * CPU emulation This paper covers years of innovation and production experience from GPU communication teams at Meta, in supporting generations of LLAMA models training and inference. Hope you enjoy reading it!

313

8 kommentarer

Wavedash loves Defold!! 🖤 Defold developers can now go from web export to live distribution, social features, achievements, leaderboards, and monetization in a few hours on Wavedash. Wild time to be building browser native games with tools like Defold and Wavedash. https://lnkd.in/gS6Szzs4

6