Kavya

Software Engineer @ SHL | Passionate about building cool stuff

[Kavya @kavya_dua_]

Managing state and retries for AI workflows is a massive headache. So I built stitch, a minimal execution engine that handles it all for you.

Just write your tasks and workflows, and the engine takes care of the rest.

Link for the dashboard in the replies.

[Kavya @kavya_dua_]

@adnaannasir_ Dodged a bullet tbh

[tanpreet 🌂 @JollyTanpreet]

introducing X-Cut

fast & lightweight chrome extension that lets you edit videos on X completely in BROWSER!

> 100% private & offline
> 100% free
> open-source

github & store link ⤵️

[Kavya @kavya_dua_]

@AbhinavXJ I like being backward deployed engineer

[tanpreet 🌂 @JollyTanpreet]

vidrush is now available for public access

single prompt → long-form video → edit it yourself or tell Rush Agent what to change

a-z in browser

live now :)

[tanpreet 🌂 @JollyTanpreet]

cursor for video editing is here.

by far one of the best features we've built at @vidrush. really happy to finally see it live.

months of work went into this one. plus a bit of back pain lol

[Kavya @kavya_dua_]

@JollyTanpreet I want one too 😅

[Kavya @kavya_dua_]

Dashboard is live here: https://t.co/50Nh0LkCBD

[Kavya @kavya_dua_]

Managing state and retries for AI workflows is a massive headache. So I built stitch, a minimal execution engine that handles it all for you.

Just write your tasks and workflows, and the engine takes care of the rest.

Link for the dashboard in the replies.

[tanpreet 🌂 @JollyTanpreet]

introducing browser-whisper v1.1

• new word-level timestamps support
• 9 new models
• OPFS cache integration
• live mic audio transcription demo
• manual download/clear models support
• 100% private, offline & open source
• new documentation

all credits to @soham_btw for this release!

[Kavya @kavya_dua_]

This is the overall architecture, very general tbh, but what I’m trying to do exactly is build a lightweight version of the currently existing orchestration tools.
Currently working on the Python SDK, and the engine(minimal and lightweight) is going to be in Go.

[Kavya @kavya_dua_]

I’ve been exploring workflow tools for agent orchestration.
There are a lot of good options, but they feel too heavy or aren’t built for AI agents.
That’s why I’m building a lightweight workflow orchestration engine in Go, designed for AI agents, along with a Python SDK.

[Kavya @kavya_dua_]

I’ve been exploring workflow tools for agent orchestration.
There are a lot of good options, but they feel too heavy or aren’t built for AI agents.
That’s why I’m building a lightweight workflow orchestration engine in Go, designed for AI agents, along with a Python SDK.

[Kavya @kavya_dua_]

Maybe they are launching something big. Chaos before singularity

[Chaofan Shou @shoucccc]

Claude code source code has been leaked via a map file in their npm registry!

Code: https://t.co/jBiMoOzt8G

[tanpreet 🌂 @JollyTanpreet]

introducing browser-whisper

audio to text transcription that runs natively in the browser

uses webcodecs + webgpu under the hood

> no backend config
> no api key required
> 100% private & offline
> just 3 lines to embed

npm i browser-whisper

github repo ⬇️

[Kavya @kavya_dua_]

@raunaksinghjoll I can't keep up with the speed of advancements 😢

[Kavya @kavya_dua_]

@abhisheknaironx Solving problems that don't exist

[Kavya @kavya_dua_]

@TheAhmadOsman I believe in the mission

[Kavya @kavya_dua_]

@cneuralnetwork My laptop is running claude opus 4.5 bro

[Kavya @kavya_dua_]

@elonmusk Is it?

[Kavya @kavya_dua_]

@gdb But it saved a lot of debugging time

[Kavya @kavya_dua_]

@NalinisKitchen I'll never be him