Stephen Fernandes (@stephennfern) header
Stephen Fernandes (@stephennfern) profile picture

Research Engineer @ CIIL building frontier intelligence for Indian Languages | also building https://xitter.gq/t.co/G8E19M11o9

Goa India
Joined April 2013
Photo by Stephen Fernandes (@stephennfern)Photo by Stephen Fernandes (@stephennfern)Photo by Stephen Fernandes (@stephennfern)Photo by Stephen Fernandes (@stephennfern)Photo by Stephen Fernandes (@stephennfern)Photo by Stephen Fernandes (@stephennfern)Photo by Stephen Fernandes (@stephennfern)Photo by Stephen Fernandes (@stephennfern)Photo by Stephen Fernandes (@stephennfern)Photo by Stephen Fernandes (@stephennfern)
๐Ÿ‡ฎ๐Ÿ‡ณ Introducing Anuvani-LID: A new state of the art in spoken language identification for Indian languages. Before AI can understand what we say, it must first recognise the language we speak. Today, the Central Institute of Indian Languages (CIIL), Government of India, unveils Anuvani-LID. It is our first open-source AI model and the first release in the Anuvani family of speech AI models. India's linguistic diversity needs models that can tell closely related languages apart and work reliably outside the lab. ๐Ÿ† Anuvani-LID ranked #1 among 16 leading AI systems in our SUITE-42 evaluation. Ahead of Gemini 3.5 Flash, Sarvam Saaras v3, OpenAI's GPT-audio, Whisper-large-v3, Meta's MMS-LID family and others. ๐Ÿฅ‡ 0.9125 mean Macro-F1 across 37 benchmark conditions (Gemini 3.5 Flash: 0.8523; Sarvam Saaras v3: 0.8372) โšก 24 percentage points ahead of Gemini 3.5 Flash on one-second speech ๐ŸŒ 0.897 Macro-F1 across all 23 supported languages, vs 0.713 for the strongest competing system ๐ŸŽ™๏ธ 22 constitutionally scheduled Indian languages + Indian English in one unified model ๐Ÿง  Trained on 54,643 hours of speech (31M+ utterances) with a compact 115.6M-parameter Conformer It's built for real-world speech: short utterances, background noise, reverberation and closely related languages. ๐—”๐—ป๐—ฑ ๐˜„๐—ฒ'๐—ฟ๐—ฒ ๐—บ๐—ฎ๐—ธ๐—ถ๐—ป๐—ด ๐—ถ๐˜ ๐—ผ๐—ฝ๐—ฒ๐—ป ๐˜€๐—ผ๐˜‚๐—ฟ๐—ฐ๐—ฒ. Released under Apache 2.0, with support for PyTorch, ONNX, NVIDIA Triton and mobile inference: โ€ข ๐—ก๐—ฉ๐—œ๐——๐—œ๐—” ๐—ง๐—ฟ๐—ถ๐˜๐—ผ๐—ป ๐—œ๐—ป๐—ณ๐—ฒ๐—ฟ๐—ฒ๐—ป๐—ฐ๐—ฒ ๐—ฆ๐—ฒ๐—ฟ๐˜ƒ๐—ฒ๐—ฟ: Production-ready serving kit for scalable, high-performance inference. โ€ข ๐— ๐—ผ๐—ฏ๐—ถ๐—น๐—ฒ ๐—œ๐—ป๐—ณ๐—ฒ๐—ฟ๐—ฒ๐—ป๐—ฐ๐—ฒ ๐—ž๐—ถ๐˜: Cross-platform toolkit for language identification on mobile and edge devices. โ€ข ๐—š๐—ผ๐—ผ๐—ด๐—น๐—ฒ ๐—–๐—ผ๐—น๐—ฎ๐—ฏ: Beginner-friendly notebook to explore, test and experiment with Anuvani-LID. โ€ข ๐—›๐˜‚๐—ด๐—ด๐—ถ๐—ป๐—ด ๐—™๐—ฎ๐—ฐ๐—ฒ: Open model weights and resources for seamless integration. ๐—™๐—ฟ๐—ผ๐—บ ๐—ฟ๐—ฒ๐˜€๐—ฒ๐—ฎ๐—ฟ๐—ฐ๐—ต ๐—ฒ๐˜…๐—ฝ๐—ฒ๐—ฟ๐—ถ๐—บ๐—ฒ๐—ป๐˜๐˜€ ๐˜๐—ผ ๐—ฝ๐—ฟ๐—ผ๐—ฑ๐˜‚๐—ฐ๐˜๐—ถ๐—ผ๐—ป-๐˜€๐—ฐ๐—ฎ๐—น๐—ฒ ๐—ฑ๐—ฒ๐—ฝ๐—น๐—ผ๐˜†๐—บ๐—ฒ๐—ป๐˜๐˜€, ๐—ณ๐—ฟ๐—ผ๐—บ ๐—ฐ๐—น๐—ผ๐˜‚๐—ฑ ๐˜€๐—ฒ๐—ฟ๐˜ƒ๐—ฒ๐—ฟ๐˜€ ๐˜๐—ผ ๐—บ๐—ผ๐—ฏ๐—ถ๐—น๐—ฒ ๐—ฑ๐—ฒ๐˜ƒ๐—ถ๐—ฐ๐—ฒ๐˜€. ๐—”๐—ป๐˜‚๐˜ƒ๐—ฎ๐—ป๐—ถ-๐—Ÿ๐—œ๐—— ๐—ถ๐˜€ ๐—ฏ๐˜‚๐—ถ๐—น๐˜ ๐˜๐—ผ ๐—บ๐—ฎ๐—ธ๐—ฒ ๐—œ๐—ป๐—ฑ๐—ถ๐—ฎ๐—ป ๐—น๐—ฎ๐—ป๐—ด๐˜‚๐—ฎ๐—ด๐—ฒ ๐—”๐—œ ๐—ฎ๐—ฐ๐—ฐ๐—ฒ๐˜€๐˜€๐—ถ๐—ฏ๐—น๐—ฒ ๐—ฒ๐˜ƒ๐—ฒ๐—ฟ๐˜†๐˜„๐—ต๐—ฒ๐—ฟ๐—ฒ. Recognising the spoken language is the first step for voice assistants, ASR, translation and accessibility. Anuvani-LID is the first milestone in CIIL's Anuvani speech AI ecosystem, with more models and open-source releases on the way. Built for India's languages. Open to the world. ๐Ÿ‡ฎ๐Ÿ‡ณ ๐Ÿ”— ๐—˜๐˜…๐—ฝ๐—น๐—ผ๐—ฟ๐—ฒ ๐˜๐—ต๐—ฒ ๐—ผ๐—ณ๐—ณ๐—ถ๐—ฐ๐—ถ๐—ฎ๐—น ๐—น๐—ฎ๐˜‚๐—ป๐—ฐ๐—ต ๐—ฎ๐—ป๐—ฑ ๐—ฟ๐—ฒ๐˜€๐—ฒ๐—ฎ๐—ฟ๐—ฐ๐—ต: ai.ciil.org/release/anuvani-โ€ฆ ๐Ÿค— ๐——๐—ผ๐˜„๐—ป๐—น๐—ผ๐—ฎ๐—ฑ ๐˜๐—ต๐—ฒ ๐—บ๐—ผ๐—ฑ๐—ฒ๐—น ๐—ผ๐—ป ๐—›๐˜‚๐—ด๐—ด๐—ถ๐—ป๐—ด ๐—™๐—ฎ๐—ฐ๐—ฒ: huggingface.co/CIIL/Anuvani-โ€ฆ ๐Ÿ“„ ๐—ฅ๐—ฒ๐—ฎ๐—ฑ ๐˜๐—ต๐—ฒ ๐—ฟ๐—ฒ๐˜€๐—ฒ๐—ฎ๐—ฟ๐—ฐ๐—ต ๐—ฝ๐—ฎ๐—ฝ๐—ฒ๐—ฟ: arxiv.org/abs/2610.20263 #AI #SpeechAI #IndianLanguages #OpenSource #LanguageTechnology #CIIL
1
1
2
392
Stephen Fernandes retweeted
Breaking news: Robert Pattinson next in line to become the Pope!
3
1
43
988
At this pace Google Deepmind will speed run the whole periodic table before Christmas RSI achieved ๐Ÿซก
2
29
Stephen Fernandes retweeted
before you disaggregate prefill and decode, do one calculation. cache transfer time = kv cache size / link bandwidth. disaggregation puts prefill and decode on separate gpu pools so they scale and schedule independently, and neither stalls the other. in exchange, every request ships its entire kv cache from one pool to the other. if that transfer time is anywhere near your prefill time, the split is eating its own win. on fast in node links it can be tiny. across nodes on a slower fabric, it can be the whole latency story. measure the link first, benchmark second.
8
5
60
2,202
People who never have launched a slurm job, people who dont even know what .einsum does in pytorch people who have never written a WGMA kernel are throwing shade at a monumental startup which India essentially needs in the coming years A startup that is handling the most critical infra of many top institutions and MNCs in the country. In Psychology this is called : " Schadenfreude " feeling intense joy and satisfaction in hoping to see others fail. enjoy the clout you clown, post this you would just be crawling back to your meaning life whoch serves no existence
How is this not a fraud? India's leading AI lab worth $1.4 billion got funded and GPU to build a sovereign model in India by the Government. They have now stopped innovating. ..and using the same GPU to sell Chinese model openly. This is absolutely rubbish and mockery.
6
19
1,044
Stephen Fernandes retweeted
People who haven't read a single research paper. Are now judging how much money it takes to build frontier models. They throw around terms like distillation and trillion parameter. Without knowing what those terms mean. This is why innovation is hard. Not only do you have to manage the models. You have to manage the public perception, including the uninformed public.
How is this not a fraud? India's leading AI lab worth $1.4 billion got funded and GPU to build a sovereign model in India by the Government. They have now stopped innovating. ..and using the same GPU to sell Chinese model openly. This is absolutely rubbish and mockery.
49
36
7
740
46,695
For all the people taking a jab at @sarvam. dig the deepest hole you possibly can when the 1T model is released you are going to need it. we are in the 2nd loop of the sarvam M cycle ๐Ÿคซ
13
1
71
5,386
Stephen Fernandes retweeted
Introducing Jev @typesafeai powered jobs-api Try it at treg.to/jobs-api - One API to any jobs data - $0.03 to run a full job hunt - Real salary & Hiring manager data - Plugin to any agent for your dream job Fully open source, 0% markup Git Repo below ๐Ÿ‘‡
189
175
123
3,237
542,863
for almost a couple of decades, this large software engineering industry put food on the table for millions of people. sadly with the advent of claude code and codex this large monolithic machinery has hyper optimized itself to remove the most unoptimized cogs in it
490
what is stripe and Grafana labs doing in this list
1
3
129
Claude opus and fable are getting insanely great at software engineering to the point where its actually scary. literally anybody can write a major component in IT that would have taken millions in engineering costs.
> it took us 4 days and 500$ in tokens > we beat SGLang by 15% > it took 6 prompts > 3 out of those 6 were: (a) "carry on" (b) "proceed" (c) "push to github" and... we did it from SCRATCH (git init) context: it has recently become a very clear industry trend to build a custom inference engine for each open source model that comes out... a paradigm made popular by a WAVE of papers that came out earlier this year, like hazy research's megakernels and MetaInfer the usual argument against traditional inference engines (like vllm / sglang) is that they, like all conventional swe products, use abstractions. the use of abstractions is synonymous with performance overhead (triton vs cuda, python vs c++, etc). combine this with increasing hardware complexity, increasing parallelization strategies, increasing kernel specialization techniques... and you can make a very clear (but very FLAWED) argument for building an engine customized purposely for one model. why flawed? it is true that traditional inference engines waste cpu time running generic abstractions. sure. but the gpu kernels (the functions run on the gpu) are almost always longer, and so the wasted cpu time is not the bottleneck. if you know your shapes, and run cuda graphs ahead of time (and you almost always do), this is NEVER a problem. so why would you build your own engine? why are people RAVING about this? so we tested it out. we built one for DeepSeek V4.1 Flash. The goal was two-fold: 1) is it feasible to do so (spoiler: yes) 2) is it useful to do so (spoiler: also yes...but for different reasons) we built it in rust (in retrospect, this did not really matter), and it took us 6 prompts and cost 500$ end to end. unshackled from working within a pre-existing codebase or inheriting code debt, we: - implemented custom kernels (GEMV, MoE, and AllReduces) specifically for DSV4.1 at our batch size - we were able to put an entire spec dec step in one graph, and verify across 7 tokens at that - we were able to mix deepseek's 4 residual streams in 1 fused kernel instead of 80 - fused the routing so that picking and running the expert happened in a single step (saving one write and one read i/o from each expert) iterating on the model was cheap to do in a custom sandbox-like engine, and we were able to go much faster than a traditional dev cycle. we were then able to port every single thing into sglang, and matched performance imo, it is this easy iteration that makes this paradigm useful. little else matters. and, ofc, we've open-sourced the engine and upstream(ed/ing) the patches to @sgl_project
2
149
Stephen Fernandes retweeted
llama.cpp turns 1.0.0 ๐Ÿฅณ It works everywhere. Open Source Will Win โœŠ
17
34
8
566
15,784
Lets Go !!!!!
Replying to @yunta_tsai
A petition is underway to reinstate Jensen and Iโ€™m signing it too
37
there's a deep sense of sadness in Silicon Valley today
2
3
183
7,052
what if we used 100% of the experts in the MoE?
32
67
11
1,321
35,383
Those who have viewed Lex's vlog featuring Jensen in Taiwan would undoubtedly recognize the profound humility, gratitude, and generosity Jensen consistently shows toward others. It is deeply painful and disheartening for me to see him depart.
NVIDIA letting Jensen go is a huge mistake, for America and for humanity. It's Apple firing Steve Jobs all over again, but much bigger, given the role of AI in defining the future of our world. I've been keeping quiet on this out of respect for Jensen's family and team, who I've met and who are amazing humans, and because I personally hate infidelity. So my heart goes out to them in this difficult personal family moment. But I can also say that Jensen is human, and makes mistakes like all of us do. Not that I matter at all, but I've known Jensen for a long time and have gotten the chance to spend a lot of time with him, behind the scenes, away from mics and cameras. I have seen him show kindness, humility, care to everyone around him, including strangers, including from the poorest most struggling parts of society, treating all the same with the same thoughtfulness and attention, because they are fellow humans and not because he has something to gain from them. That shows rare character, to maintain humility and humanity through all of it. Another data point is what people who know him well have said privately to me about Jensen over the years. I've never had anyone tell me anything but how awesome he is or how much they love him. Probably over a thousand people at this point. This includes hundreds of engineers, CEOs, colleagues, friends, etc. Also, since there was some spy talk. Jensen is as proud an American as it gets. I'm in shock frankly at all of this, including the mass hysteria around it. I love America, and I think NVIDIA letting Jensen go is terrible for America. I stand with Jensen. Love you all โค๏ธ
1
111
Stephen Fernandes retweeted
Me refusing to sleep because i didn't get enough me time after work
811
88,845
4,034
556,717
9,769,355
with better infra support we need to enable confidence and trust in most teams to use preemptible instances
everyone wants a faster kernel. nobody wants to hear that their gpu is idle most of the time
1
60
Stephen Fernandes retweeted
i never cuda expected this
60
553
32
9,192
170,671