Solopreneur | indie hacker

Kochi, Kerala
Joined April 2010
Include

Only show posts containing:

Exclude

Hide posts containing:

Time range
-
Minimum likes
Replying to @pradx
Use gemma4 models if you need to run it locally. For generic off the shelf solution checkout gemini rag.
9
Krishna Sangeeth KS retweeted
We release LFM3 today, starting with a family of 2B-class models
17
29
2
344
8,142
support for malayalam and indic languages would be great.
41
Krishna Sangeeth KS retweeted
An early attempt at getting @googlegemma 's Embedding Gemma 2 to paint using its shared understanding of text and image. 30 ms(around 30 images per second) per image on the Apple Neural Engine on a base Mac M1 pro. generates an image at the end of every word. Trained for about 12 hours on an A100. It is still undertrained. But, with enough data, images can get a lot better.
12
7
2
137
6,476
Krishna Sangeeth KS retweeted
Maya-2 voice model weights are officially live! 🇮🇳🇮🇳 The frontier voice model for India is now completely open-weight. • Outperforms ElevenLabs v3, Sarvam Bulbul v3, and Cartesia Sonic 3.5 • 1.5B+ people covered across all 28 states and 8 union territories • SOTA voice synthesis across 11 official languages • 20 native speakers pre-loaded, ready out of the box • Best-in-class zero-shot voice cloning Built by a team of five from the land of languages. Go build and have fun!
195
355
44
3,434
136,652
Krishna Sangeeth KS retweeted
Unlimited OCR 2 is here, and it's open source. 📖 ✨ Longer documents can mean a lot more attention computation. Our new SR-SWA mechanism keeps that in check by condensing text-visual attention interactions into a fixed-size set of working references. Here's what it delivers: - Document parsing at virtually any length, within a finite context window - Nearly constant per-step decoding latency and memory footprint as documents grow longer - State-of-the-art performance on MPDocBench and MarathonDocBench, covering documents from dozens to hundreds of pages Check out the model here ↓
62
115
61
1,067
135,312
Krishna Sangeeth KS retweeted
I regret to inform you that nobody cares about your potential. Potential is nothing without execution. When you're young, everyone cares about your potential. You get used to focusing on it. But as you get older, no one cares about your potential. Just your delivery against it.
267
1,967
84
15,471
344,016
Replying to @vashinthewild
:D Also a huge security concern no, that a model changes its behaviour by changing few neuron’s.
1
1
15
Cool stufff
i opened up an AI's brain tonight with @_lyraaaa_'s neuron explorer qwen 3.5 4b has 294,912 neurons push just one and it becomes someone else here's who i found 🧵
1
1
238
Tech is ready and available for disruption. Personalised 1:1 education is now technically feasible, whether multi decade systems and processes would be ready to change is the biggest bottleneck.
1
25
Me and @_TheIndianRaj_ built a diagnostic evaluator for an @AIBoomi kochi hackathon recently. It’s a full duplex voice based evaluator that tests student’s knowledge assimilation through socratic Q&A method youtube.com/watch?v=tsAnUGd6… We used Qwen for ASR, Silero for VAD, turns3.2 for turn detection, GPT as reasoning LLM and Phonon as TTS. Entire thing was orchestrated with Pipecat. We were able to build a full duplex system that can scale within the timespan of a hackathon. Goes to show how easy it is becoming to build voice AI systems now.
1
7
386
Krishna Sangeeth KS retweeted
SGLang and vLLM are racing 😭. There’s no weekend for inference guys.
24
53
5
922
45,306
Krishna Sangeeth KS retweeted
This is what Crome saying “this is your last opportunity to fix the mess you’ve created, else you’re out.” Pretty clear they will be out. We’re talking about an eng team unable or unwilling to run a RCA and publish it, when everyone assumed they could bugzilla.mozilla.org/show_bu…
19
65
18
2,014
167,291
Replying to @gnurjnj
They have this DSC business which seems to be the major revenue source as it’s mandatory to be used within government portals. But the thing is there is already a rival product called Adhaar E-sign which is more convenient for users. DSC token needs renewals and is expensive while Adhaar E-sign is cheaper. If a new government comes and decides to switch to E-sign then the DSC business is gone. It makes more business sense for them to have continued their TLS cert business, if executed well it was a huge opportunity because it’s hard to become a CA.
1
46
Krishna Sangeeth KS retweeted
Can you use EmbeddingGemma 2 for Speech-to-Text? This experiment uses EmbeddingGemma 2’s unified audio-text vector space to match speech against text via vector similarity. The result? Transcribing an hour of audio in about 13 seconds, locally!
56
158
21
1,887
84,621
Replying to @shobhitic
I don’t really see the need for DSC business also when there is already Adhaar E-Sign. Not enough people currently asks why should i buy and renew a DSC token YoY to make a company profitable. If there is enough pressure to migrate the various government portals to accept Adhaar e-signed docs, then i don’t know what happens to this company. That’s a legit risk. So they shouldn’t have left the TLS business IMO. Accept the mistake and move on by creating better processes. It’s definitely a matter of prestige to be a CA if you do stuff in the right way.
1,172