Lead Web 3 Engineer | Devrel | Podcaster | @web3wikixyz, @coverantxyz | @Moca_Network / @animocabrands | Advisor | Army Veteran | Man on many missions
Dubai, United Arab Emirates
Joined October 2021
- Tweets11.9K
- Following951
- Followers1.2K
- Likes12.6K
Arman retweeted
‼️ After PHP2Shell comes Node2Shell, a critical bug found by OpenAI's Codex Security that can let attackers run code in Node.js apps.
It was successfuly demonstrated on the official Node.js 24.22.0 image.
Node 20, 23 and 25 (all EoL) get no fix.
Arman retweeted
We were able to fine tune an open weight model (Qwen3.5-9B) on proprietary internal data at @coinbase, and it out performed frontier models, at much lower cost and speed, for a specialized task (fraud prevention in this case).
I think we'll see more of this pattern inside companies. The frontier models are amazing generalists. But increasingly you'll see specialist models out perform on more narrow tasks, either working on teams, or by themselves.
We're making it easy to fine tune models internally for our employees. We even have a skill for it in our internal harness that makes it easy to spin these up and start training them:
"/fine-tune-and-serve-llm I want to fine-tune Qwen for payments risk review. help me get started"
I imagine we'll have some backend services that could be replaced entirely by this new architecture.
Our new docs are live at docs.air3.com/
We've reorganized everything around how you build with AIR (@Moca_Network ), including examples, diagrams and actual use cases, so it's easier to find what you need and build with us.
Arman retweeted
We instituted an official AI Writing Policy at @Clay.
Massive thanks to @sophiebits on our engineering team who wrote this.
Originally this was just for eng, but other teams found it so helpful we expanded it company-wide.
Here are the four guiding principles:
1. You must stand behind every idea and sentence
It is your responsibility to make sure that the entire document is representative of your own thoughts before you share it.
2. Writing is thinking
Spending time on the writing process teaches you more about your topic. If you circumvent this process, you will walk away with a poorer understanding of the subject matter.
3. More time should be spent writing a document than consuming it
If you generate a document from a short prompt then ask your readers to go through the longer output, you are disrespecting their time. They can talk to ChatGPT themselves if they want to.
4. Longer is not better
AI makes it easy to generate long docs, and it loves padding them with sentences that say nothing. If you're producing docs from a short prompt, consider just sharing the prompt.
You can read (and borrow) the full policy below. Hopefully we all communicate a bit more clearly now :)
Yep. Although there is no x4 fee layer when you're building with issuer from scratch.
Like if you go and build one with Rain or Bridge, you basically reducing 2 fee layers
Just the beginning of shipping The RIGHT products for the market
WEB3 evangelists Arman Mamyan @primarchdev, has introduced OpenCasp-KYC, an open-source reference designed to help Armenian crypto-asset service providers translate KYC requirements into code.
We’re proud to see his expertise contributing to shared resources for Armenia’s growing Web3 ecosystem. Well done, Arman🙌🏻
github.com/armanmamyan/OpenC…
Just read the news from @Shopify that they’re moving away from @reactnative and going back to native app development. Quite a questionable move I'd say.
Their main arguments seem to be that AI has reduced the cost of maintaining two separate codebases, while React Native introduces additional complexity around performance optimization, framework updates, and development tooling. (seems a very strange argumentation). What's strange is that @expo already solves a significant portion of these problems.
From EAS Build and OTA updates to development clients, simulator support, and web compatibility, Expo makes the entire development experience considerably smoother. Reading about Shopify's challenges, I have a feeling they weren't fully leveraging Expo and ended up building and maintaining tooling that the ecosystem already provides.
From my own experience building with React Native, especially when working at the edge of the modern stack, where so many libraries are web-first and native polyfills are often missing, React Native is a gem. It's not perfect, and sometimes you still need native code. But being able to share business logic, iterate quickly, ship updates without waiting for app store approvals (where permitted), and access native capabilities when needed is an incredibly powerful combination.
AI making it easier to maintain two codebases doesn't automatically make maintaining two codebases a better architectural decision.
I'm sure lot of people will consider me a newbie on React Native field (been building on it for only 3 years), so I welcome to hear your thoughts on this.
The real story to tell here is the missing transparency for supply-chain and custody vulnerabilities in many "web3" products.
Customers thought they used an Avici product but ended up using Rain, thinking it was a fully self-custodial solution.
Stopping the bad guys with Cloudflare: 189 malicious requests blocked or challenged in the last month #cloudflare
@defyneric Dispute resolution is the most difficult topic. every decision can be disputed as well.
From your POV what dispute resolution should look like? What would satisfy the market at least 75-80% ?