
👨💻 Staff Engineer at @brexhq, leading platform. I love writing about software, teams, productivity, and more. Almond croissant expert 🥐
Vancouver, Canada
Joined April 2022
- Tweets2.6K
- Following353
- Followers1.1K
- Likes8K
Konstantin retweeted
I want to share a preview of something I've been working on called Exact. It's a new Agent Native app framework written in Rust. It's extremely fast and tries to let you build exactly the experience you want with no compromises.
Everyone sees rain in Vancouver, i see the ski season arriving early ⛷️ hope that does not get too warm.
1
298
Giving tax breaks to ppl who barely pay taxes, amazing policy 😂 gives signaling without losing actual tax revenue.
In Budget 2026, we’re backing the Canadian artists who tell our stories.
That means if you’re a Canadian artist, you won’t pay tax on up to $45,000 of income derived from your creative work.
1
102
This is just ridiculous… you’re alienating young professionals who are trying to save us for this ridiculous real estate situation that your own government nurtured into a disaster.
National Bank says you need $265,618 to qualify for the average Vancouver home. Eby says you’re rich so the new 22.5% tax bracket kicks in at $265,545.
Thanks for playing.
112
Odd to see non-10x take, but completely agree here. I spent a lot more time with agents actually investigating different parts of product and user feedback.
the right way to use model capabilities is not to ship 10x more features to prod
it's to spend more time understanding your users, trying experiments, building prototypes, learning about things you don't understand so that you can ship things that actually work
1
300
Konstantin retweeted
Introducing 𝚒𝚜𝙴𝚟𝚎𝚗 / 𝚒𝚜𝙾𝚍𝚍, but with @typesafeai Jev + @aisdk
Now available on NPM: 𝚗𝚙𝚖 𝚒 𝚓𝚎𝚟-𝚎𝚟𝚎𝚗-𝚘𝚍𝚍
Sometimes i wish REST only had GET and POST.
PS Spicy take that QUERY is a step to the wrong direction 😂
1
1
253
Konstantin retweeted
I’ve got two senior-to-staff level engineering roles opening up on my team at @phantom. We’re hiring across US/EU time zones.
If you’re an amazing mobile or backend engineer and want to work at the intersection of AI + trading, DM me or reply below.
Hot take - the only thing models need are a proper API with an OpenAPI spec.
Everything else is like 95% sugar on top of it that models can just generate themselves.
I was not expecting things to go this way, but I think MCPs are better than CLIs for most integrations.
The models have gotten much better at tool calling, we can defer tools & MCP is now stateless.
If you need to compose/filter data, add params like query to your MCP tools.
195
The thing I’m really interested is if we ever gonna get OTAs on native apps.
AGI may be closer in this timeline
7
332
Konstantin retweeted
Here's an offer.
Reach out and talk to me before deciding to move from React Native to native.
Totally free: you'll get the advantage of my 15 years of doing mobile app development, both native and RN (not to mention my 32 total years of coding experience). I'll give you up to 3 hours of my own time to talk to your whole dev team.
I promise I will absolutely not just try to convince you to use RN, despite that being my specialty for the past decade. I will have nuanced opinions and quite honestly may just tell you that you should move to native, once I've heard the situation.
I just want people to have the right info before making a decision.
The dream, I’m curious if we already coined the agent version of XSS tho. This feels like a real problem to me in fully agentic end-to-end system when request comes from a non-authed actor.
The night, the pharmacy, the street...
A sandbox, a function, a build, a server. They're all expressions of the same underlying compute infrastructure on Vercel.
Fluid Compute is one system that creates the machine your workload needs, on demand.
vercel.com/blog/fluid-comput…
1
388
There is no war in Ba Sing Se 😂
We unfortunately have decided that we cannot continue providing access to our models through Cursor and are ending our partnership. It boils down to trust and we’ve asked that this takes effect on November 12 to give you some time to plan.
Many have used the GPT models through Cursor and here are options we know should work in the future:
- We will continue to allow using your own OpenAI API key and similarly will continue to provide access through our IDE extensions for Cursor.
- We will keep working with the broadest range of tools and harnesses, some of which are OSS, but also many many closed-source ones.
We are as committed as ever to continue supporting developers and the flourishing ecosystem of tools, harnesses and products. We will also continue to invest in our own open-source initiatives and believe in broad optionality for developers.
You can read more about our decision in the blog:
openai.com/index/our-decisio…


















