Oleksandr Zinevych

🦸‍♂️Engineering Director #nodejs #ruby at @Avenga_global. 👨‍💻 Tweeting about Tech, IT, Management ✍️

[Oleksandr Zinevych @a_zinewicz]

Трішки поділюсь тут своїм новим проєктом 😎 https://t.co/soWmDxUx1q

[Theo - t3.gg @theo]

If you recognize this image, I hope you enjoy your upcoming retirement

[Oleksandr Zinevych @a_zinewicz]

@rtheoryxyz @ThePrimeagen Or when it hits during hot-fixing production

[ThePrimeagen @ThePrimeagen]

we are about to hit 1 9 of availability while coding is largely solved

[Oleksandr Zinevych @a_zinewicz]

*smth* claw like back in js frameworks boom era 😌

[Andrej Karpathy @karpathy]

Bought a new Mac mini to properly tinker with claws over the weekend. The apple store person told me they are selling like hotcakes and everyone is confused :)

I'm definitely a bit sus'd to run OpenClaw specifically - giving my private data/keys to 400K lines of vibe coded monster that is being actively attacked at scale is not very appealing at all. Already seeing reports of exposed instances, RCE vulnerabilities, supply chain poisoning, malicious or compromised skills in the registry, it feels like a complete wild west and a security nightmare. But I do love the concept and I think that just like LLM agents were a new layer on top of LLMs, Claws are now a new layer on top of LLM agents, taking the orchestration, scheduling, context, tool calls and a kind of persistence to a next level.

Looking around, and given that the high level idea is clear, there are a lot of smaller Claws starting to pop out. For example, on a quick skim NanoClaw looks really interesting in that the core engine is ~4000 lines of code (fits into both my head and that of AI agents, so it feels manageable, auditable, flexible, etc.) and runs everything in containers by default. I also love their approach to configurability - it's not done via config files it's done via skills! For example, /add-telegram instructs your AI agent how to modify the actual code to integrate Telegram. I haven't come across this yet and it slightly blew my mind earlier today as a new, AI-enabled approach to preventing config mess and if-then-else monsters. Basically - the implied new meta is to write the most maximally forkable repo and then have skills that fork it into any desired more exotic configuration. Very cool.

Anyway there are many others - e.g. nanobot, zeroclaw, ironclaw, picoclaw (lol @ prefixes). There are also cloud-hosted alternatives but tbh I don't love these because it feels much harder to tinker with. In particular, local setup allows easy connection to home automation gadgets on the local network. And I don't know, there is something aesthetically pleasing about there being a physical device 'possessed' by a little ghost of a personal digital house elf.

Not 100% sure what my setup ends up looking like just yet but Claws are an awesome, exciting new layer of the AI stack.

[Oleksandr Zinevych @a_zinewicz]

Working on some telegram bot for parents for a while. Maaaybe it time to share my progress here and start building in public? 😏

[Oleksandr Zinevych @a_zinewicz]

@mattpocockuk Great article! Useful insights! Hope one day this can be easily integrated in some big legacy project and work well.

[Matt Pocock @mattpocockuk]

https://t.co/OINxN3w3e0

[Oleksandr Zinevych @a_zinewicz]

@rbuckton Sorry to hear that... Good luck with all the new opportunities for you ;)

[Oleksandr Zinevych @a_zinewicz]

As usual, the most significant attack is the dumbest reason behind it: https://t.co/jusj2E1o1i

[Matt Pocock @mattpocockuk]

Coding assistants have made for the case for TypeScript 10x stronger.

- AI agents use TS's type errors as an extra feedback loop
- AI agents need handholding to investigate runtime errors
- The cost of writing types has dropped to zero

Please, don't use .js for anything

[Oleksandr Zinevych @a_zinewicz]

Yet another day, when we still have our jobs and are not replaced by AI 🥲

[Oleksandr Zinevych @a_zinewicz]

AI will not replace us; we will be replaced by people who use AI

[GREG ISENBERG @gregisenberg]

LEAKED internal memo from CEO of Shopify @tobi around AI

10 quick takeaways i have after reading it:

1. a subtle but huge reframe: “hire an AI before you hire a human.”
2. AI is now a baseline expectation at shopify. hiring filters will probably favor ai-fluent candidates at shopify and other companies.
3. AI agents are now treated like teammates, not tools.

4. prompting is now a core skill. top performers will be top prompters.

5. AI usage is now measured. kinda wild. probably a business idea there to build the lattice for AI usage.

6. AI-first prototyping is the new standard. shipping speed will probably 10x even at a $100B company like Shopify.

7. org charts blur, headcount planning now includes bots, not just bodies.

8. AI literacy is the new coding literacy. prompting, contextualizing, or evaluating ai output is become mandatory.

9. AI is now a core layer in the software stack. not a plugin. not an add-on. ai sits beside infra, backend, frontend, and design. the best teams will be the ones who treat it like infrastructure.

10. tobi’s memo screams one thing: more impact per person. shopify is early to this, but i bet this will hit every major company over the next 12-24 months.

[Oleksandr Zinevych @a_zinewicz]

@andrii_sherman So real 🤪

[Oleksandr Zinevych @a_zinewicz]

Vibe coding is a shitty idea at least for now

[Oleksandr Zinevych @a_zinewicz]

All you need to know about the incredible capabilities of modern AI assistants :)

[Theo - t3.gg @theo]

Grok 3 is here and it fails the "hexagon ball bouncing" test spectacularly

[Oleksandr Zinevych @a_zinewicz]

The backward compatibility between Deno and Npm is incredible. For now, no significant difficulties or problems. 😌

[Matt Pocock @mattpocockuk]

I have officially read enough of it to say:

This book is great btw

[Daniel San @dani_avila7]

Deepseek running locally and privately for autocompletion in VSCode! 🙌

In less than a minute, I'll show you how to download Deepseek-coder and set it as the autocompletion model in VSCode.

You’ll need to use @ollama to download the model and CodeGPT to select it as the autocompletion model.

Enjoy the best models running locally with https://t.co/2TwiyF71HB :)

[Oleksandr Zinevych @a_zinewicz]

@lukedepulford @deepseek_ai It’s not smth I wouldn’t expect 🫠