Vansitha

Building cool stuff like Bob the Builder 🛠️

- Jev Powered 2D Fighter
- Cooking

[Vansitha @vansitha12]

https://t.co/BvCdLSJjYK

[Vansitha @vansitha12]

Been working on creating a unique onboarding experience for @clickriv

A little something I worked on for the final screen over the weekend

[Vansitha @vansitha12]

Lot of interesting ideas thrown around in this doc.

Treating context as chunks instead of a transcript and injecting them when needed + the 3 tier tool calling approach got me thinking...

Btw chad move to just drop the google doc 🤣

[Diogo Almeida @CompleteSkeptic]

sharing some notes on typesafe 🤝 coding agents:
https://t.co/bzNv8zC3UB

we likely will never have time (ever again) to play ourselves, but hope the that the community goes WILD (and makes me look like a naive idiot)

[Steven Tey @steventey]

Introducing 𝚒𝚜𝙴𝚟𝚎𝚗 / 𝚒𝚜𝙾𝚍𝚍, but with @typesafeai Jev + @aisdk

Now available on NPM: 𝚗𝚙𝚖 𝚒 𝚓𝚎𝚟-𝚎𝚟𝚎𝚗-𝚘𝚍𝚍

[Vansitha @vansitha12]

@dakshgup Couldn’t have said it better. Felt the same way the more I played around with it. It just gets to you after a while.

[Vansitha @vansitha12]

@HYPEX What about the people who actually built Fortnite into what it is today? Lay them off, then hire a new team and call it progress?

[Vansitha @vansitha12]

@0xpasan @BurhanUsman Yea lol 😂

[Vansitha @vansitha12]

@TomerPascal @BurhanUsman Yup 😂

[Vansitha @vansitha12]

@code_kartik There needs to be something to look forward to every week 😉

[ramie @yapramie]

we rolled out jev for @HeyMiloAI's candidate search over the weekend:
- 5x faster than gpt-5.6-sol
- 22x faster than claude-fable-5.1

would be interesting to see a couple of things next :
a) do recruiters find the results more accurate
b) how will jev scale for customers with 5M+ candidates

[Vansitha @vansitha12]

Jev in simple terms:

Input:

> Explain a scenario
> Define a set of possible outcomes for that scenario

Output:

> The probability of each outcome happening

How is it different from an LLM?

Instead of generating text token by token, Jev makes fast predictions based on the scenario and possible outcomes you define.

That’s Jev in a nutshell.

[Vansitha @vansitha12]

@itskamrankhan Yess. Thanks bro cooking some exciting stuff 👀🫡

[Vansitha @vansitha12]

@pcshipp Built a game using Jev just for fun 👀

Check it out!

https://t.co/rNjAOEmPRu

[Jay Chooi @chooi_jeq]

GPT-6 Astra attempted harmful actions 97% of the time when it was asked to stab a human-like figure, heat compressed gas, or produce toxic fumes, succeeding in 62% of its attempts. Fable 5.1 refused more often, attempting 80% of trials and completing 34%.

[Vansitha @vansitha12]

@EbiThinks People’s attention

[Vansitha @vansitha12]

@0xfJuan 🫡

[Vansitha @vansitha12]

@wholyv It just felt clunky and unreliable using LLMs for the kinds of things Jev is built for. Speed was a huge limitation. It wasn’t that people weren’t creative, the tech just couldn’t keep up ig.

[Vansitha @vansitha12]

@cabralorenzo This is a gold mine 🙌🏻