Prakash Palnati

Software Development ➡️ Books ➡️ Badminton

[Prakash Palnati @itzppal]

And we should…

[Prakash Palnati @itzppal]

We’ve been treating LLMs as black boxes.

“J-space” feels like cracking open a tiny window into the control room.

Possibly to explain how information comes together before the next token.

https://t.co/4SMpCdNq1I

[Prakash Palnati @itzppal]

🎯

[Prakash Palnati @itzppal]

“How LLMs work” — in pakka telugu 😉

[Manjunath @MIrukulla]

GPT ni adugu "STRAWBERRY" ane word lo enni letters unayi ani. Watch it fail! Counting raaka kaadhu, dheeni venaka chala peddha reason undhi. Adhi teliyali ante, meeku Tokenization gurinchi teliyali saami...!

Antha peddha katha emi kaadhu le saami.

[Prakash Palnati @itzppal]

How practical ! Golden advise.
Applies to any engineer who wants to tinker around or build something ..

BTW, @MIrukulla love your choice Telugu memes 😀.

[Manjunath @MIrukulla]

Projects cheyali, ideas raavatle ana prathi saari tension padakunda I just follow this framework, and I always end up with a production-grade project.

Motham thread chaduv, you'll understand it completely. 🧵👇

[Prakash Palnati @itzppal]

@MIrukulla Really liked your breakdown..
More power to you for especially breaking it down using Telugu memes 💙..

[Prakash Palnati @itzppal]

@tarak9999 @saitejaathoti You are such a joy .. 🤩

[Prakash Palnati @itzppal]

@saitejaathoti @pavansahas @tarak9999 Lovely …

[Prakash Palnati @itzppal]

@arpit_bhayani 🤭🤭

[Prakash Palnati @itzppal]

@miryalasrikanth Thanks doc ..

[Prakash Palnati @itzppal]

@SquintNeon Protect her at any cost ..

[Prakash Palnati @itzppal]

Never by chance , always on purpose!!

This quote on my coffee-mug motivated me a bit more today.

[Prakash Palnati @itzppal]

@arpit_bhayani Congratulations!

[Prakash Palnati @itzppal]

@LiamHarreys @GergelyOrosz Nice one 😀😀

[Prakash Palnati @itzppal]

So cool. Love it when de-facto is challenged and solved

[Nishkarsh @contextkingceo]

We've raised $6.5M to kill vector databases.

Every system today retrieves context the same way: vector search that stores everything as flat embeddings and returns whatever "feels" closest.

Similar, sure. Relevant? Almost never.

Embeddings can’t tell a Q3 renewal clause from a Q1 termination notice if the language is close enough.

A friend of mine asked his AI about a contract last week, and it returned a detailed, perfectly crafted answer pulled from a completely different client’s file.

Once you’re dealing with 10M+ documents, these mix-ups happen all the time.

VectorDB accuracy goes to shit.

We built @hydra_db for exactly this.

HydraDB builds an ontology-first context graph over your data, maps relationships between entities, understands the 'why' behind documents, and tracks how information evolves over time.

So when you ask about 'Apple,' it knows you mean the company you're serving as a customer. Not the fruit.

Even when a vector DB's similarity score says 0.94.

More below ⬇️

[Prakash Palnati @itzppal]

Nice pivot bro 😎, smart 👍🏻

[Perplexity @perplexity_ai]

Announcing Personal Computer.

Personal Computer is an always on, local merge with Perplexity Computer that works for you 24/7.

It's personal, secure, and works across your files, apps, and sessions through a continuously running Mac mini.

[Pratyush Kumar @pratykumar]

📢 Open-sourcing the Sarvam 30B and 105B models! Trained from scratch with all data, model research and inference optimisation done in-house, these models punch above their weight in most global benchmarks plus excel in Indian languages.

Get the weights at Hugging Face and AIKosh. Thanks to the good folks at SGLang for day 0 support, vLLM support coming soon. Links, benchmark scores, examples, and more in our blog - https://t.co/DcCG3zlN8p

[Prakash Palnati @itzppal]

@theBuoyantMan Agree 💯

[Prakash Palnati @itzppal]

@GergelyOrosz Anthropic almost always seems to be the first in the market for most problems.But they might loose when things consolidate.

Microsoft has advantage of having Homebase (PC). If they do a decent job , people wouldn’t for cowork.
I did the same with codex against Claude-code.

[Prakash Palnati @itzppal]

@siddharthb_ 🤣