prathosh ap

The tech industry convinced an entire generation of developers that they can skip the math. They are wrong. You cannot build foundational architecture with a "GenAI in 5 days" bootcamp. Real engineering requires staring at the equations until they make sense. If you actually want to build state-of-the-art models, you cannot skip the math. No fluff, no quick-bytes. Just the mathematical foundations of Deep Generative Modeling. Here is a course I designed to put everything you need to learn.

@anandbs_291 @anushilansamiti Probability theory

@anushilansamiti @prathoshap Are there any prerequisites or subject knowledge needed before going through the playlist?

ATTENTION AI ENTHUSIASTS.... One of my most favorite teachers, Dr Pratosh AP (@prathoshap) has joined X. The first time I came across his videos was sometime in 2025 when I watched his Mathematical Foundations of Gen AI course. Fantastic course ++ youtube.com/playlist?list=…

Just closed the lid on the meta-review/AC cycle for ICML 2026. It’s easy to look at a spreadsheet of 4s and 5s and hit "Accept," but the real job is finding the "Reject" papers that actually have merit. I spent today diving into rebuttals and reviews to override committee noise where the science was actually solid. Yes, I accepted papers with reject scores, too. This is where humans still score above AI. Agents can process the "Knowledge" and flag the gaps, but "Judgment", recognizing a nuke in a knife fight, is a uniquely human play. All of our fates (I'm an author, too) are now officially sealed in the cosmic servers of OpenReview. Good luck to all of us.. #icml26 #metareview #research #openreview

We translated state-of-the-art machine learning concepts into one of the world's most ancient and structurally perfect languages. You can watch the full interview here: Full Broadcast - youtube.com/watch?v=v7OoAv…

Fools call it a "dead language"....... Meanwhile, we are debating the nuances and applications of AI, neural networks, and the future of artificial intelligence on national television (DD News) exclusively in classical Samskrutam (Sanskrit). A language engineered with mathematical precision is the perfect medium to discuss computation.

@deburdened Of course. It's a necessity for becoming a good researcher, but not sufficient. You obviously know the difference.

@prathoshap You can know the equations and still fail to build anything that composes. You need a representation where correctness is easier to check than error is to generate.

The tech industry convinced an entire generation of developers that they can skip the math. They are wrong. You cannot build foundational architecture with a "GenAI in 5 days" bootcamp. Real engineering requires staring at the equations until they make sense. If you actually want to build state-of-the-art models, you cannot skip the math. No fluff, no quick-bytes. Just the mathematical foundations of Deep Generative Modeling. Here is a course I designed to put everything you need to learn.

@anushilansamiti Thank you for your kind words. Your suggestions will be duly incorporated :)

@BlizzzzBastard Most welcome. You know where to find me.

@prathoshap If possible can I catch you on some free day for some more ways of thinking things. Would love even few mins ig

Incredible set of Lectures on Sastra and Vendatic Philosophy by Dr Prathosh AP(IIsc Prof) Made a playlist on some of the few I found. youtube.com/playlist?list=…

AI can replicate your code in weeks. It cannot replicate what your system learned by running. That gap is the moat. Here is what I think is valuable and why It's important to act now: aravindjayendran.com/writing/irredu…

@HriteshMaikap no no. i just randomly sampled. You can see the entire course notes in the link provided.

@prathoshap Asking out of curiosity, by any chance is the first slide about policy gradient derivation?

@AnkitNa83620147 probability theory and vector calculus

@prathoshap Sir , what's the prerequisites (like math level and others) to start this course??

Everything is now public and open-source for those who care about the inner workings of generative architecture. Full Course playlist - youtube.com/playlist?list=… Lecture Notes & Assignments - drive.google.com/file/d/1YTciDI… Github Repo - github.com/Chandan-IISc/I…

I originally taught this curriculum for the IIT Madras BS program. We bypassed the slides and went old-school: writing out every single equation, derivation, and proof by hand. Then, we implemented every concept in PyTorch from scratch. What it covers: • Diffusion Models • RLHF (PPO, DPO) • GANs & VAEs • Auto-Regressive & Language Models

@SriSank0505 Sure, will do.

@prathoshap It will be great if this can be extended to Sama veda as an immdt priority and Atharvana Veda as next priority. thank you

Namaste X! This is my first post here - I'm humbled to release my app: VedaVaaNi. It is an interactive learning platform designed to help you practice Rig Veda and Krishna Yajur Veda chanting with absolute phonetic and prosodic (Chandas) precision.

@K_V_I_N will do eventually once it matures.

@prathoshap Could you please make it open source and host the repository somewhere so it can be added via Obtainium thus eliminating the dependency on Play Store?

@nandakumarm FYI - I have personally. taken the consent of Acharyas .

@prathoshap If such an app was required, acharyas of thrimatas would've sanctioned it already. Your disclaimer won't stop anadhikaris. It's upto you anyways.

@JunoVortex @ravii_mishra Slow recitation is already there.

@ravii_mishra @prathoshap Probably, we can have two modes, in a learning mode, we could have a slow recitation, a gap for repetition, or a pause with subtitles in (any language of choice option) so one could understand before proceeding further..

@Ekk_Bhartiye It should work with these specs. May I know your country of residence?

@Ekk_Bhartiye Version? I

@Yuyutsu101 @b1tchless_chow @nandakumarm absolutely.

@b1tchless_chow @nandakumarm @prathoshap Guru is must