For me Da Nang is the perfect place to work from. Everything is quickly accessible, great food and the best coffee in sea. And in Feb-March weather is comfortable under 25 celsius
OK fine no gatekeeping. Da Nang has amazing beach clubs and they are all free entry. Menu prices are reasonable too ($5-10 for an entree).
The photo is this post is from Blush Beach Club (hidden gem) and here are some of the other top tier ones:
Esco, Maia, Paradise, Kala Kala, East West Brewing.
Vietnam culture is all about coworking so it's perfectly normal to have a laptop out a beach bar, cafe, restaurant etc. Most places even have extension cables sprawled out for people to use. Compare this to cafes back home in LA that block off the outlets so you can't stay too long.
Building on @Polymarket? Then you should know that every user action is onchain
The problem: transactions are unreadable without context
I built an open-source tool + guide how to:
* Subscribe to Polymarket events via RPC
* Decode transactions into human-readable messages
🚨 I was just targeted in a sophisticated phishing attempt that almost got me.
But I got the scammer on a live call (video recording below), strung him along, and trolled him with Kim Jong Un gay porn while dissecting his $3k/month malware kit.
Buckle up, this gets wild. 🧵👇
I started with React Native because the startup we were building had separate apps in Kotlin and Objective-C. The apps had a bunch of bugs and were slow, and took long to update whenever new features were released.
I decided to convert our React app into RN, and in just 2 months we shipped apps that were better in every way.
Apple today took a shot at React Native and other cross-platform frameworks.
I have thoughts.
First, the quote in question, from their keynote (around the 40 minute mark):
"Some other frameworks promise the ability to write code once for Android and iOS.
And that may sound good, but by the time you've written custom code to adapt each platform's conventions, connected to hardware with platform-specific APIs, implemented accessibility, and then filled in functionality gaps by adding additional logic and relying on a host of plugins, you've likely written a lot more code than you'd planned on.
And you are still left with an app that could be slower, look out of place, and can't directly take advantage of features like Live Activities and widgets.
Apple's native frameworks are uncompromisingly focused on helping you build the best apps."
@dcinvestor@Apple I played with it today, and the only place it feels a bit too much is the Control Center. They can easily add more blur and darken the background more to create the illusion of greater distance. They do that in app library, and it looks more than fine
By too complex, I was referring to the API, it requires a lot of smaller blocks to be composed manually. I think having one place to configure everything through parameters has less mental overhead when re-reading the code.
By limiting, I was referring to the fact that other frameworks already handle multiple providers and have a good abstraction. For example, switching from model it just a small change, and they will handle different specifics like how to handle the reasoning output on a model.
Overall my experience was with Effect AI I had to look into internals to understand something. With Vercel AI, I just had to look at the type and I would directly understand how to achieve my goal.
I'm pretty sure Effect team can also achieve a good state for this package (I love how effect/sql turned out for example), but for now I just have a simple wrapping service and the experience was pretty good.
@schickling@EffectTS_@tylerangert I have migrated the other day from it to a custom wrapper service of vercel AI. I found effect version too complex and limiting. The worst was the dependency on the third party tokenizer which gave me a lot of headaches. Vercel AI just worked for every usecase
@EffectTS_ Ngl, I fed it a lot of context and an example implementation in a PR from @imax153
I have to update some code style, but I'm pretty sure 80% will remain from AI
But here is the code https://t.co/J8aP4IJYJe
Claude 4 + Agent mode has just helped me to write a Circuit Breaker and Request Pool with @EffectTS_
It is mind-blowing how good it is. The Agent wrote the module, documentation, tests and fully integrated it in the current system.
Another day, another journo.
Today it’s a guy named Ali Breland. He’s been spamming our members for months, because the oligarch Laurene Powell Jobs is paying him to write a hit piece on our startup society.
Let’s talk about the specific angle he was tasked with writing about, and how our hard numbers completely disprove his journo rumors.
Then let’s go through the general context of why he was assigned to write this fake piece in the first place.
We'll end with the obvious observation that an Atlantic that abolished the police and bombed the Middle East actually has no moral standing whatsoever — and certainly cannot stop the free people of the world from pursuing our God-given right to self-determination by peacefully building new societies.
HARD NUMBERS > JOURNO RUMORS
First, the specifics. Ali was assigned to harass us using a common journo tactic one might call “Trouble in Paradise.” That’s when the journo finds a few people you had to let go, puts their worst-sounding complaints under a magnifying glass, and then uses those distorted outliers to tar the entire organization.
Now, for years, the way journos did this was to baselessly accuse people of rayyycism. They could usually find some party claiming you had secret bigotry, and failing that they’d just manufacture an accusation themselves. And indeed that’s exactly the line Ali pursued on a different company, as you can see from his past unsolicited DM to me in 2023:
Of course, it wasn’t just this DM. Given that he’s a journo, Ali has falsely accused many things of being rayyyycist over the years. A partial list of things he’s tarred as Nazi-adjacent include NFTs, e/acc, Telegram, Gamestop, AMC, DeSantis, Elon, Reddit, and even Silicon Valley itself.
But calling everything rayyyycist isn’t as effective anymore, and it’s particularly ineffective on another brown guy like me, so Ali had to come up with a new angle: rather than secret bigotry, he’d accuse us of being secretly unhealthy! Behold the message he’s been spamming to our community members since at least Jan 2025:
Note the part about “mold”. That’s meant to sound bad. Except it’s completely misleading. Because mildew in Southeast Asia is as common as sand in the Sahara or frost in Canada. It’s just a common building maintenance issue in Singapore’s humid climate, and present even in beautiful new places like Nanyang Technological University:
With that context, you can understand what actually happened: we rented several buildings in late September, we heard about spots of mildew a few weeks later in October, and we hired a top Singapore firm to remediate it out of an abundance of caution. Their professional opinion was that it was just cosmetic with zero health implications. I’ve attached their letter below:
In addition to this expert opinion stating no health issue, we looked at the instrumental record of air quality measurements across the facility from our dehumidifiers and air purifiers. The absolute worst reading we got was well within the normal range of Los Angeles air, and of course the average air quality was much better than LA on average.
So: that’s a fatality for Ali’s fake story. Because when journos accuse you of racism, there isn’t a geiger counter for bigotry. There’s no numerical way to prove yourself innocent of their false accusations. But because there is a geiger counter for air quality, we can prove ourselves innocent of the charge that there was any health issue.
Essentially: against mere journo rumors, we have hard numbers. So Ali will have to give up his attempted cosplay as a molecular epidemiologist, and return to his original job of accusing people of rayyycism in his capacity as a fake news journalist.
By the way, this is exactly how Elon knocked down a fake NYT story that claimed Tesla cars ran out of charge on the side of the road. He proved the NYT journo was faking the story, because he had the instrumental record.
Similarly, we too have the instrumental record of air quality measurements from all our IoT devices. And now that they’ve been informed that both expert opinion and on-the-ground quantitative measurement detected no health issues whatsoever, any attempt by Ali and his oligarchic bosses at the Atlantic to defame our startup society could be considered actual malice…with all the consequences that entails for their media corporation.
But why do these oligarchs want to defame us in the first place? For that let's go to tweet #2:
This playground uses hardcoded mapping. Initially the library was designed for a realtime transformation pipeline and LLM are way too slow for this use case. But one can easily pass the decoded object to LLM, and it would work just fine with a right prompt
Available interpreters https://t.co/XTlkbi9I0I
@hklst4r Additionally, we have an interpreter that makes it human-readable. Devs don't need to learn specifics of a contract to show transactions into their app, the library will abstract all of them
https://t.co/jdMWLB5kmZ
@hklst4r Thank you for spending time to check it out. This is a library intended for developers who want to integrate it directly into their app. It's free and one can build tenderly like features inside their own app using this library, but it's not limited to this use case.
Retro Funding: Dev Tooling is backing the devs who make building on the Superchain easy for all.
@3loop_io by @nastyarods & @ferossgp is a powerful open-source tool that makes onchain transaction data human-readable - no closed APIs required.
Why that matters ↓