Kabya Basu

7.2K posts

Kabya Basu

Kabya Basu

@kabyabasu

MLOps/LLMops Engineer

Paris, France Katılım Mart 2010
406 Takip Edilen988 Takipçiler
Kabya Basu retweetledi
anand mahindra
anand mahindra@anandmahindra·
Five years ago, the team at @GalaxEye were students at IIT Madras, chasing what many would have called an “outrageously ambitious” dream. Yesterday, they launched Mission Drishti, the world’s first OptoSAR satellite. There’s a simple but powerful lesson here: leapfrogging the world begins with a leap of faith. By combining optical and radar imaging into a single platform, they didn’t just improve on what existed, they reimagined what was possible. They’re showing us that the way to take a place on the global stage, is not by catching up, but by changing the game. They’re my #MondayMotivation
GalaxEye@GalaxEye

Separation Confirmed! The world's first OptoSAR Satellite is now in space. Made in India for the world. Go Drishti! Go @GalaxEye! Go India!

English
59
345
2.7K
106.8K
Kabya Basu retweetledi
Hindu Voice
Hindu Voice@HinduVoice_in·
#Kolkata A powerful TMC leader grabbed this playground and built fencing for promoting. Because of this, children couldn't play for years. Today, after the landslide victory, local youths with the BJP workers recovered the playground.
English
147
5K
20.8K
381.6K
Kabya Basu retweetledi
Swapna Kumar Panda
Swapna Kumar Panda@swapnakpanda·
Nandan Nilekani: India doesn't need to lead in building AI models. But must lead in its usage. The toddlers like Anthropic is now $1T company, OpenAI is $850B. Thanks to leaders like him, entire Indian IT ecosystem is not even $1T. Not a single company is even 10% worth of Anthropic still being for >50 years in business. TCS - $94B (56 years old) Infosys - $49B (43 years old) HCL - $34B (50 years old) Wipro - $21B Tech Mahindra - $15B L&T - $14B They will always say we are mass recruiters. But they won't say why engineers are paid peanuts. They won't say why work-life balance is the worst. They won't say why employees are so outdated. They will just pick that thing which is easy, which is simple.
Swapna Kumar Panda tweet media
English
125
199
745
43.5K
Kabya Basu retweetledi
R𝛼m🦅
R𝛼m🦅@rambuilds_·
As an AI Infrastructure Engineer. Please learn: - GPU/VRAM fundamentals, quantization & batching - vLLM / TensorRT-LLM / inference optimization - KV caching, speculative decoding & token throughput - Distributed training basics (DDP/FSDP/DeepSpeed) - Model serving & autoscaling - Vector DB retrieval pipelines - Prompt caching & cost optimization - Observability for LLM apps This is what production AI teams actually care about.
English
9
79
650
17.6K
Kabya Basu retweetledi
freeCodeCamp.org
freeCodeCamp.org@freeCodeCamp·
Fine-tuning turns a general LLM into a reliable specialist for your particular domain and tasks. And in this course, Tatev teaches you how to fine-tune your LLMs. You'll learn what fine-tuning means, different methods, how to work with massive models on your home workstation, and more. freecodecamp.org/news/how-to-fi…
freeCodeCamp.org tweet media
English
4
51
419
14.7K
Kabya Basu retweetledi
freeCodeCamp.org
freeCodeCamp.org@freeCodeCamp·
AI agents are emerging as a powerful tool for helping devs get things done. And in this tutorial, @itsTarun24 walks you through building your own agent using Gemini in Python. Your AI Study Planner agent will be able to remember your conversation history & use a search tool to access relevant online resources. freecodecamp.org/news/how-to-bu…
freeCodeCamp.org tweet media
English
7
51
425
11.8K
Kabya Basu retweetledi
PyTorch
PyTorch@PyTorch·
Want to train LLMs on longer contexts without re-engineering your entire systems stack? Introducing AutoSP — the first compiler-based solution that automatically optimizes LLM training for long contexts. Under the hood, AutoSP applies a series of compiler passes that trigger sequence parallelism, paired with a curated activation-checkpointing scheme tailored for long-context training. It's integrated directly into DeepSpeed, so enabling long-context training is just a config change away. No more rewiring your stack to push context lengths. Read the blog to learn more 🖇️ pytorch.org/blog/introduci…@AhanGupta13, Zhihao W., Neel Dani, @toh_tana, Tunji Ruwase, @_Minjia_Zhang_ #PyTorch #DeepSpeed #AutoSP #OpenSourceAI
PyTorch tweet media
English
3
21
120
15.3K
Kabya Basu retweetledi
Captain Insight
Captain Insight@CaptainInsightX·
The man who killed the $10,000 GPU myth. He did it alone, from Bulgaria, with one C file. 🤯 >Meet Georgi Gerganov. >Bulgarian developer. Nobody had heard of him. >In March 2023, Meta’s LLaMA model leaked online >Within days he wrote a single C file >Called it llama.cpp >It ran a full AI model on a MacBook. No GPU. No cloud. >The entire AI industry said you needed $10,000 GPUs to run LLMs 🔥 >He proved you didn’t. On a laptop. Alone. >Also built whisper.cpp ~ same thing for voice AI > His code is the foundation of Ollama, LM Studio, and GPT4All >107,000+ GitHub stars. Fastest open-source AI project to hit 100K ever. 🚀 >In 2026 Hugging Face hired his entire team >Still ships code. Still open source. Still free. Every time you run AI locally, you’re running his work. Absolute Legend 🐐
Captain Insight tweet mediaCaptain Insight tweet media
English
119
977
8.5K
260.1K
Kabya Basu retweetledi
Aritra 🤗
Aritra 🤗@ariG23498·
[Hugging Face Machine Learning Club India] It is with immense gratitude and exictement I announce the next speaker for our virtual event, @sarahookr! The date, time, details on how to join will be shared in due time. Share this announcement with you friends, and help make this a great event! 🤗
Aritra 🤗 tweet media
English
4
15
140
5.6K
Kabya Basu retweetledi
DeepLearning.AI
DeepLearning.AI@DeepLearningAI·
Eli Schilling's workshop at AI Dev 26 focused on Memory Engineering and Context Engineering, and shared a working mental model of both disciplines and hands-on exposure to building a memory-first agent harness using @Oracle AI Database, LangChain, and Tavily. "An AI Agent without memory is just autocomplete with ambition. Give your agents the memory they deserve so they can stop forgetting and start learning." Explore the session's repo: github.com/oracle-devrel/…
DeepLearning.AI tweet media
English
4
17
95
7.8K
Kabya Basu retweetledi
freeCodeCamp.org
freeCodeCamp.org@freeCodeCamp·
The best way to understand how LLMs work is to build one yourself. In this handbook, Wisamul teaches you how to create a language-specific model step by step. You’ll go from raw text data to a working chatbot you can customize and learn how these models work under the hood. freecodecamp.org/news/how-to-bu…
freeCodeCamp.org tweet media
English
3
91
595
20.2K
Kabya Basu retweetledi
Utkarsh
Utkarsh@utk7arsh·
I thought robotics was for PhDs and billion dollar labs. Then I found this repo where NVIDIA open-sourced the entire stack for physical AI. Brain. Body. Physics. Simulation. Free. I wrote the full breakdown and what projects you can start building today to get ahead.
Utkarsh@utk7arsh

x.com/i/article/2048…

English
14
77
618
65.4K
Kabya Basu retweetledi
freeCodeCamp.org
freeCodeCamp.org@freeCodeCamp·
Most RAG systems retrieve information, but they don't learn from it. In this tutorial, @dannwaneri shows how to add a knowledge reflection layer that links new docs to existing ones and synthesizes insights. Along the way, you’ll build a self-learning RAG system that improves as you ingest more data. freecodecamp.org/news/how-to-bu…
freeCodeCamp.org tweet media
English
2
54
331
11.1K
Kabya Basu retweetledi
vLLM
vLLM@vllm_project·
🎉 Congrats to @NVIDIAAI on Nemotron 3 Nano Omni — a 30B hybrid Transformer-Mamba MoE (3B active) that unifies vision, audio, video, and text in a single reasoning loop. 256K context, FP8 / NVFP4 quantization, open weights. Day-0 support in vLLM — tool calling, reasoning, and efficient video sampling for long-video workloads, verified on NVIDIA GPUs. 🔗 vllm.ai/blog/nemotron-… 🔗 recipes.vllm.ai/nvidia/Nemotro…
vLLM tweet media
NVIDIA AI@NVIDIAAI

Meet Nemotron 3 Nano Omni 👋 Our latest addition to the Nemotron family is the highest efficiency, open multimodal model with leading accuracy. 30B parameters. 256K context length. 🧵👇

English
11
50
412
33.2K
Kabya Basu retweetledi
DeepLearning.AI
DeepLearning.AI@DeepLearningAI·
We're opening the doors at AI Dev 26 today, and celebrating with a little something for everyone: AI Prompting for Everyone, a new course taught by @AndrewYNg, is now live! 🎉 Learn how to get accurate answers, better writing, and more useful outputs from the AI tools you already use — whether you're an engineer, a marketer, a salesperson, in finance, or anything in between. Enroll today: hubs.la/Q04dFX7B0
English
6
47
278
19.2K
Kabya Basu retweetledi
THE SKIN DOCTOR
THE SKIN DOCTOR@theskindoctor13·
No need to come back. Indians must understand the difference between Bharat, the oldest unbroken dharmic civilization since the Bronze Age, and India, a modern nation-state governed by a democratic system since 1947. Your love and gratitude for Bharat should not be weaponized as a guilt trip to pull you back to India, where merit is devalued, corruption is normalized, adulteration is rampant, civic sense is poor, pollution is pervasive, babushahi stifles efficiency, appeasement and freebies shape policy, and mobs dictate terms to democratically elected governments. Bharat stays with you. Wherever you go, the civilization goes with you. If you want to preserve and carry forward that civilization, practice dharmic righteousness wherever you are and uphold its values through your conduct, work, and integrity. You don't have to be in India for it.
Sridhar Vembu@svembu

Open letter to Indians in America. -- Dear brothers and sisters from Bharat: Like I did 37 years ago, you arrived in America with no money but with a good education and cultural heritage from Bharat. You achieved outstanding success. America was good to us. For that we must remain grateful - gratitude is our Bharatiya way. Yet today, a significant number of Americans, may be not the majority but not too far from it either, believe that Indians "take away" American jobs and our success in America was unfairly earned. You may think the next election will fix this, but your choice would be between people who hate our Bharatiya civilisation and people who hate civilisation itself. That is the "hard right" vs "woke left" battle. You are mere bystanders to that conflict. Meanwhile there is one thing that is true now and will be true in the future: the respect Indians command world-wide will substantially depend on the fortunes of India herself. If India remains poor, the woke left will give us moral lectures with pity and the hard right, different moral lectures with scorn ("hellhole") and we must not confuse either with respect. Respect in today's world, along with prosperity and security, comes from one source: a nation's technological prowess. India produces sufficient brain power to achieve that prowess but alas we exported so much of that talent, particularly to America. As we develop that prowess in India, our civilisational strength will assert itself. As difficult as it is for many of you to contemplate this, please come back home. Bharat Mata needs your talent. Our vast youthful population needs the technology leadership you gained over the years to guide them towards prosperity. Let's do it with a missionary zeal. Respectfully Sridhar Vembu

English
341
2.1K
10.1K
262K
Kabya Basu retweetledi
freeCodeCamp.org
freeCodeCamp.org@freeCodeCamp·
LLMs are powerful tools on their own - but incorporating agents can really boost their capabilities. In this tutorial, @djibril_mg walks you through building an autonomous AI agent using LangChain and LangGraph. You’ll learn about summarizing data into text, creating tools the agent can call, and setting up LangGraph nodes to orchestrate workflows. freecodecamp.org/news/how-to-bu…
freeCodeCamp.org tweet media
English
5
43
288
12K
Kabya Basu retweetledi
vLLM
vLLM@vllm_project·
vLLM v0.20.0 is here! 752 commits from 320 contributors (123 new). 🎉 Highlights: DeepSeek V4, Hunyuan v3 preview support, CUDA 13 / PyTorch 2.11 / Transformers v5 baseline, FA4 as default MLA prefill, TurboQuant 2-bit KV (4× capacity), vLLM IR foundation. Thread 👇
vLLM tweet media
English
23
80
670
67.6K
Kabya Basu retweetledi
Vaishnavi
Vaishnavi@_vmlops·
RAG Evaluation & Testing in Production (Offline + Online) Interview Preparation Playbook for Model Evaluators, AI QA & SDETs for LLMs drive.google.com/file/d/1nvKRSs…
Vaishnavi tweet media
English
3
96
559
24.5K
Kabya Basu retweetledi
Oier Mees
Oier Mees@oier_mees·
I am thrilled to host @QuanVng (Co-Founder of @physical_int ) as our Guest Spotlight at @ETH! I will be starting the afternoon with a lecture on Generalist Robot Policies and how to train VLAs, before handing over to Quan to close out the session. Quan has been a central figure in the push toward large-scale pre-training for generalist robot policies. From co-leading the RT-X and Open X-Embodiment projects at Google DeepMind to his current work at Physical Intelligence, his efforts in scaling physical AI are defining the next frontier of robotics.
Oier Mees tweet media
English
1
16
187
9.3K