Hariom Chaudhary

@DipankarDas011 What you uses for hardware old laptop or arduino, rasberrypie

Lets see how it goes

@DipankarDas011 👍

My linux environment after a new machine migration ☺️

@akshayktwt @Apple Crucial mx500

Content creation 🎥 | Community work 🤝 | Open source 💻 And my #MacBook’s 256GB storage just gave up 😅 Looking for a fast & reliable external SSD. What do you folks use/recommend? 👇💾 #opensource #community #contentcreation @Apple

@DipankarDas011 nice view nice work

2025 year wrap-up 👉Travelling to Kerala and Rameswaram 👉Getting hands on with Kubernetes in production setting 👉Experience with prod. RBAC design, Rate limiting, System design and billing 👉Custom Statemachine for my background workers. 👉kubmin product x.com/DipankarDas011…

@nitishfy Nice work

2025 was a year of growth, setbacks, and quiet resilience. > got graduated this year from college > received bonus from my org as an intern for a good performance > got my first full time job > got shoutout at KubeCon EU and KubeCon NA for contributions to Argo CD project > built multiple features e2e for Argo CD project and got featured multiple times in release blogs > travelled to London and Edinburgh > spent almost one month in US visiting atlanta, sf, la, vegas > started content creation and crossed 850+ subscribers on youtube > did a one week trip to vrindawan with parents (best) > lost a couple of dearest friends but that’s okay > learnt to play piano > took skincare and fitness seriously > invested time and money into learning > read 3 books (need to do better) I am not sure how 2026 is going to turn around but i hope it ends up making me grow both personally and professionally.

𝗛𝗼𝘁 𝘁𝗮𝗸𝗲: Your RAG system doesn't need chunking. In fact, adding it might be killing your performance. Before diving into complex chunking techniques, start by asking yourself this simple question: 𝘋𝘰𝘦𝘴 𝘮𝘺 𝘶𝘴𝘦 𝘤𝘢𝘴𝘦 𝘢𝘤𝘵𝘶𝘢𝘭𝘭𝘺 𝘳𝘦𝘲𝘶𝘪𝘳𝘦 𝘤𝘩𝘶𝘯𝘬𝘪𝘯𝘨? 𝗪𝗵𝗲𝗻 𝘆𝗼𝘂 𝗗𝗢𝗡'𝗧 𝗻𝗲𝗲𝗱 𝗰𝗵𝘂𝗻𝗸𝗶𝗻𝗴: • Your documents are small and fit comfortably within your LLM's context window (like FAQs, product descriptions, or social media posts) • Each document covers a single, focused concept • You're working with simple, structured data where the entire document is the relevant unit • Speed is critical and your documents are already optimally sized In these cases, chunking might actually hurt performance here. 𝗪𝗵𝗲𝗻 𝘆𝗼𝘂 𝗗𝗢 𝗻𝗲𝗲𝗱 𝗰𝗵𝘂𝗻𝗸𝗶𝗻𝗴: • Your documents are large and contain multiple distinct topics • You need precise retrieval of specific facts within longer documents • You're working with diverse document types (PDFs, HTML, Markdown) • You need to balance retrieval precision with contextual richness The goal is to find the "chunking sweet spot" - chunks that are small enough for precise retrieval but complete enough to give your LLM the full context it needs. Save this guide and explore our full ebook on Context Engineering: weaviate.io/ebooks/the-con…

@MuthaNagavamsi Im 26 next month and want a full time job before come my next birthday

im 38. does life get better after 40?

@DipankarDas011 Same to you bro

Happy diwali 🪔 to all 🎉

Linux commands a-z desk mate is back in stock!✨ Save yourself time and energy by keeping a reference on your desk🐧 Grab yours: shop.sysxplore.com

@DipankarDas011 github.com/JaKooLit/Fedor…

@hariomtwt Use hyperland as it uses wayland Don't go to i3 Most apps do work

Actually Hyprland has became very stable and even screensharing and other things work even slack works Yes finally better replacement for i3 based on X11 If you are the one who uses i3 I would recommend you to switch to hyprland as x11-server support in desktop env is going away

@boredabdel RUN 🏃‍♂️at gym some weight training🏋‍♂️ Code👨‍💻

What's your weekend plan? Mine are: - Run 🏃 - Tea 🫖 - Code 💻 Regardless, may your weekend sports be injury-free, your tea kettle be full of water, and your code be free of AI 🙌

How to Be More DISCIPLINED: - Thread -

@iamnarhari Congratulations Narhari !!

Passed CKAD! :)

Natural ability alone isn't enough. Without consistent effort, discipline, and dedication, talent is wasted.

Kubernetes Interview Question buff.ly/4gKQtIh

Join CNCF Ambassador Halil Bugol as he explored OpenCost for the first time! hubs.la/Q03k95H_0

@itsrajputamit @TransUnion Congrats Bro

It's been more than a week, I joined @TransUnion as Developer DevOps and leading with my contributions to native identity pillar under TrueAudience marketing wing. The past three months have been a whirlwind of learning, growth, and self discovery also didn't thought that I would be making some tough choices in joining a new organisation. Grateful forever ✨ PS: Thought of sharing it with my twitter fam before anywhere else also got bunch of interview questions that I have been asked and would be sharing it across to help the community, soon.

i'm literally frustrated with @reliancejio at this point. we're paying 700+ rs/month for the 30 mbps plan for Jio Air Fiber & this is the speed we're getting since the past 1-2 weeks. - customer service not picking up at all (tried 6-7 times since yesterday) - no option to register a complaint via the app basically, no support from their team & the connection has never been constant since we started the service! very bad connectivity & customer service tbh! thinking of other options now. location: Shimla, HP

@prateek_0041 RISC V is also good option

if you are still in college, you MUST take microprocessors classes VERY seriously. If the teachers are teaching 8086, when you come back home, read the x86 or ARM architecture in-depth. I did the same. You are going to be insanely good with low level engineering.