Vinayaka
banner
vinayakakv.com
Vinayaka
@vinayakakv.com
Software Engineer & Photographer
Just tried SORA a few mins back, felt crappy
January 19, 2025 at 9:11 AM
Kyle Hill made a video on this — youtu.be/aFlromB6SnU

It is a part of a playlist Half life histories - youtube.com/playlist?lis...
Demon Core - The True Story
YouTube video by Kyle Hill
youtu.be
January 11, 2025 at 11:14 AM
In my latest article, I explore one such case in the field of deployment management. Check the article here - vinayakakv.com/blog/you-hav...

2/2

#devops #k8s #software
Dear Sir, You Have Built Kubernetes
A tale of how a simple deployment script turned into a monster that could rival Kubernetes.
vinayakakv.com
December 19, 2024 at 10:26 AM
Move fast, break things, **don't burnout**

Take a leave and rest!
December 19, 2024 at 7:34 AM
Omg, it is turning out so cool!

You built a new language with LSP and all the tooling, right?
December 18, 2024 at 2:26 AM
First it was `Object.fromEntries`, now it is `Object.groupBy` - Low how the ecosystem is getting more ergonomic!

Though I'm forced to be on the latest LTS of Node (sometimes even the non-LTS) for good.
December 17, 2024 at 2:53 AM
Ah yeah. I read yesterday Ollama now supports structured outputs. If they support function calling, It might open the door to a ChatGPT-like local UI.

ollama.com/blog/structu...
Structured outputs · Ollama Blog
Ollama now supports structured outputs making it possible to constrain a model's output to a specific format defined by a JSON schema. The Ollama Python and JavaScript libraries have been updated to s...
ollama.com
December 13, 2024 at 3:36 AM
Ollama?
December 12, 2024 at 10:00 AM
Imagine this appearing on web workers, node, bun and deno one day
December 11, 2024 at 3:25 PM
`window.ai`, it is now available in Chrome canary hidden under a flag I think
December 11, 2024 at 3:24 PM
Atleast (()=>{})() is more manageable
December 9, 2024 at 3:54 AM
Reminds me of `:(){:|:&};:`
December 9, 2024 at 3:54 AM
Redux, ask Dan!
December 7, 2024 at 3:59 PM
You bend the spacetime...!
December 5, 2024 at 4:17 PM
maybe

```js
const goodSourceFetchInput = await getSourceFromLlm(input)
const someGoodSource = await fetch(goodSourceFetchInput)
// Same as below...
```

RAG is basically LLMs identifying which function to call, us calling and providing the result to it as context
December 4, 2024 at 12:25 PM
Good to see you building things for TS, everything in ML is so much focused around Python!
December 3, 2024 at 3:53 PM
C!
December 3, 2024 at 9:21 AM
Nice hills!
November 27, 2024 at 4:27 PM
Yeah, we use it in prod and it works great!
November 24, 2024 at 4:44 AM