Looking for a male replacement in a fully furnished 4 BHK Duplex in Indiranagar near metro station (Gated Society with a pool view room)
More details: shorturl.at/MhtsM
English
Vishant Batta
1.9K posts
Vishant Batta
@Vishant_Batta
Building AI tools, bullish on the future of privacy-first AI | Currently @writesonic | Ex @unacademy | IIT Varanasi | Chardikala
Punjab → Bengaluru, India Katılım Temmuz 2017
1.4K Takip Edilen801 Takipçiler
Vishant Batta retweetledi
"Hydra reasons asynchronously, speaks and listens simultaneously, and preserves emotion"
That seems like ADHD ngl
Jokes aside, great stuff @kamath_sutra 🙌
Sudarshan Kamath@kamath_sutra
OpenAI's S2S preview is polished but it still thinks in steps. Speech → text → model → text → speech. That's not how humans converse. Introducing Hydra. A native speech-to-speech model that doesn't wait for turn-taking, doesn't flatten emotion into text, and doesn't break when you interrupt it mid-sentence. Hydra reasons asynchronously, speaks and listens simultaneously, and preserves emotion because it never leaves the audio domain. It's still in beta, but the shift is obvious. If you want early access, the link is in the comments. Here's a preview of what that looks like -
English
Vishant Batta retweetledi
1/
I’ve been building a bunch of things lately after getting back to coding in the last few weeks.
But one problem kept sitting in the back of my mind: agent workflows were still too wasteful on code search/retrieval.
English
@divz_i_am Childhood Trauma as a service
Gives you details, resources and proofs of childhood traumas according to your past.
English
Soon we’ll have people making up childhood trauma for funding.
Alice Bentinck@Alicebentinck
Fundraising in London vs. San Francisco? Apparently, being asked about childhood trauma is universal... Konstantin raised $4.5M in two weeks across both locations, led by Northzone and Initialized. We discussed the difference in speed, questions, and processes. Our full conversation in the first comment.
English
@tr1ckastley @ThineAI Would love to try it! Have been experimenting with something like this on my own
English
Not the kinda data poisoning enterprises will like @AnthropicAI
English
All systems are go. @NotionHQ custom agents are finally here!
PS. This is such a beautiful war room 😅
Notion@NotionHQ
Introducing Custom Agents. The AI team that never sleeps 🌙 They’re autonomous, built for teams, and easy for anyone to build. Give them a job, set a trigger or schedule, and they'll get it done 'round the clock.
English
I gave my claw bot eyes and ears. It now watches and listens as I work - learning from me in real time. Never felt this powerful.
How are you solving context for Openclaw beyond MCPs?
English
@shantanugoel Notion does this pretty well. Plus unlimited Claude.
English
I've been using Capacities as my main note taker over the last year and it works great! Except the AI integration is a bit lackluster.
I mainly want to be able to chat with my all my notes without having to provide context manually, and find out information as needed. Trying out obsidian with copilot plugin and it's great for this!
Going to try out khoj as well soon to integrate a few more of my resources (like karakeep).
English
@chintanpuggalok @tush_tr604 thank god it didn't roast in Punjabi
English
@dravishakatoch Basic: Try reducing social media, should do most of the trick
Should do: Read something that interests you everyday. Keeps the momentum good for longer reads
Can try: Listen to audiobook and textbook at the same time - engaging multiple senses reduces distraction.
English
@SimonHoiberg For not-so-big vector DB, even a simple numpy file works efficiently.
English
I just brutally cut another 3rd-party dependency.
Our Vector Database: Pinecone.
Pinecone used to be an essential building block in Aidbase's RAG system.
It would store tens of millions of individual knowledge points from the custom models our users are training on their content.
Our replacement?
A simple Postgres database using pgvector.
Hosted on Hetzner.
Here's what changed:
1. Speed: Lower latency on our actual workloads
2. Cost: A fraction of the monthly bill
3. Flexibility: Normal SQL + vectors in one place
4. Control: No vendor lock-in, easier scaling
5. Simplicity: Fewer services, fewer failure points
The myth:
You need "specialized" tools for production-grade RAG.
The reality:
- You don't need the scale you think.
- Optimize where it actually hurts.
- Keep ownership of your data.
Fancy tools isn't a strategy.
Outcomes are.
In this case:
pgvector won.
English
@shobhitic My friend once went on the right path, he was challaned ₹500.
English
just like how you know you're on the right path in a game when bad guys keep showing up that you gotta beat
you know you are on the right path IRL when fear keeps showing up after beating a previous fear
the thing you fear is actually the direction to move to
random thought
English
AI builders are experiencing imposter syndrome at scale.
Watching everyone else's launches and achievements while questioning your own progress, despite being in the same learning curve as everyone else.
You're not behind. You're exactly where you should be.
English