Satyabrata pal

Machine learning architect during the week and on weekends I write about Deep learning👨‍🏫, Fitness🏋️‍♂️ and Photography

[Satyabrata pal @TheCodingProjec]

I'm thrilled to announce my newest YouTube tutorial! Dive into key NLP concepts, tackle real-world datasets, and attempt your first kaggle competition🚀 Watch here: https://t.co/weaRUIK8GO 🎉 Remember to like, subscribe, and ring that notification bell! 🔔

[Satyabrata pal @TheCodingProjec]

@amazonIN it is rather a continuous trend that orders are marked as delivered or delivery attempted by delivery agent even if it wasn’t delivered. This is so frustrating and unacceptable.

[Andrej Karpathy @karpathy]

nanoGPT - the first LLM to train and inference in space 🥹. It begins.

[Satyabrata pal @TheCodingProjec]

@amazonIN why is your service so pathetic. This is second time in a row this week that my order’s status has been misrepresented by delivery agent and despite that the status of the product is the same as it was yesterday. On top of it there is no option to call the customercare

[Satyabrata pal @TheCodingProjec]

@amazonIN your service has degraded over time. This has been the 7th or 8th time my order was marked as delivered but was not handed over to me. On top of this the process of raising complaint is so utterly frustrating.

[Satyabrata pal @TheCodingProjec]

@lexfridman @drmichaellevin it’s amazing to see how @lexfridman churns up such interesting content one after another.

[Lex Fridman @lexfridman]

Here's my conversation with Michael Levin (@drmichaellevin) about the nature of intelligence in biological systems, including unconventional & alien intelligence, agency, memory, consciousness, and life in all its forms here on Earth and beyond.

It's here on X in full and is up everywhere else (see comment).

Timestamps:
0:00 - Introduction
0:44 - Biological intelligence
9:17 - Living vs non-living organisms
14:30 - Origin of life
18:15 - The search for alien life (on Earth)
51:19 - Creating life in the lab - Xenobots and Anthrobots
1:04:21 - Memories and ideas are living organisms
1:18:02 - Reality is an illusion: The brain is an interface to a hidden reality
2:03:48 - Unexpected intelligence of sorting algorithms
2:29:26 - Can aging be reversed?
2:33:17 - Mind uploading
2:51:57 - Alien intelligence
3:06:52 - Advice for young people
3:13:21 - Questions for AGI

[Andrej Karpathy @karpathy]

As a fun Saturday vibe code project and following up on this tweet earlier, I hacked up an **llm-council** web app. It looks exactly like ChatGPT except each user query is 1) dispatched to multiple models on your council using OpenRouter, e.g. currently:

"openai/gpt-5.1",
"google/gemini-3-pro-preview",
"anthropic/claude-sonnet-4.5",
"x-ai/grok-4",

Then 2) all models get to see each other's (anonymized) responses and they review and rank them, and then 3) a "Chairman LLM" gets all of that as context and produces the final response.

It's interesting to see the results from multiple models side by side on the same query, and even more amusingly, to read through their evaluation and ranking of each other's responses.

Quite often, the models are surprisingly willing to select another LLM's response as superior to their own, making this an interesting model evaluation strategy more generally. For example, reading book chapters together with my LLM Council today, the models consistently praise GPT 5.1 as the best and most insightful model, and consistently select Claude as the worst model, with the other models floating in between. But I'm not 100% convinced this aligns with my own qualitative assessment. For example, qualitatively I find GPT 5.1 a little too wordy and sprawled and Gemini 3 a bit more condensed and processed. Claude is too terse in this domain.

That said, there's probably a whole design space of the data flow of your LLM council. The construction of LLM ensembles seems under-explored.

I pushed the vibe coded app to
https://t.co/EZyOqwXd2k
if others would like to play. ty nano banana pro for fun header image for the repo

[Sebastian Raschka @rasbt]

Implemented Olmo 3 from scratch (in a standalone notebook) this weekend!
If you are a coder, probably the best way to read the architecture details at a glance: https://t.co/wF8PkoDuBe

[Unitree @UnitreeRobotics]

Embodied Avatar: Full-body Teleoperation Platform🥳
Everyone has fantasized about having an embodied avatar!
Full-body teleoperation and full-body data acquisition platform is waiting for you to try it out!

[Satyabrata pal @TheCodingProjec]

@ForrestPKnight Management thinks AI is writing 90% code but in reality it's only 10% in some cases. The same management folks equipped with the latest and greatest coding IDE spin up a prototype in 2hrs just to prove that we are not using AI effectively.

[Satyabrata pal @TheCodingProjec]

@ForrestPKnight AI solution architect here- In my experience if it's an existing application and if someone is trying to add and integrate new features then AI is able to assist me in 10%-20% of my code.

[Andrej Karpathy @karpathy]

Transforming human knowledge, sensors and actuators from human-first and human-legible to LLM-first and LLM-legible is a beautiful space with so much potential and so much can be done...

One example I'm obsessed with recently - for every textbook pdf/epub, there is a perfect "LLMification" of it intended not for human but for an LLM (though it is a non-trivial transformation that would need human in the loop involvement).

- All of the exposition is extracted into a markdown document, including all latex, styling (bold/italic), tables, lists, etc. All of the figures are extracted as images.
- All worked problems get extracted into SFT examples. Any referenced made to previous figures/tables/etc. are parsed and included.
- All practice problems are extracted into environment examples for RL. The correct answers are located in the answer key and attached. Any additional information is added as "answer key" for a potential LLM judge.
- Synthetic data expansion. For every specific problem, you can create an infinite problem generator, which emits problems of that type. For example, if a problem is "What is the angle between the hour and minute hands at 9am?" , you can imagine generalizing that to any arbitrary time and calculating answers using Python code, and possibly generating synthetic variations of the prompt text.
- All of the data above could be nicely indexed and embedded into a RAG database for later reference, or maybe MCP servers that make it available.

Then just as a (human) student could take a high school physics course, an LLM could take it in the exact same way. This would be a significantly richer source of legible, workable information for an LLM than just something like pdf-to-text (current prevailing practice), which simply asks the LLM to predict the textbook content top to bottom token by token (umm - lame).

As just a quick and crappy example of synthetic variations of the above example, GPT-5 gave me this problem generator (see image), which can now generalize that problem template to many variations:

- When the time is 11:07 a.m., what is the degree measure of the angle between the hands? (Answer: 68)
- Determine the angle in degrees between the clock’s hands at 4:14 a.m.. (Answer: 43)
- What angle do the clock hands form when the time reads 11:47 a.m.? (Answer: 71)
- At 7:02 a.m., what angle separates the hour hand and the minute hand? (Answer: 161)
- At 4:14 a.m., calculate the angle made between the two hands. (Answer: 43)
- What angle is formed by the hands of a clock at 4:45 p.m.? (Answer: 127)
- What is the angle between the hour and minute hands at 8:37 p.m.? (Answer: 36)
(infinite practice problems can be created...)

[Curiosity @CuriosityonX]

Tiny Moon 'Daphnis' creating giant waves in Saturn's Rings.

[Satyabrata pal @TheCodingProjec]

41years of wait and India has come far since then.

[वरिष्ठ प्र बंधक @Bebasbankerz]

Congratulations and proud moment for India, after 41 years, in the space, Shubhanshu Shukla, a new chapter in the Indian Space Mission. 👏👏👏🌌🛸🇮🇳🇮🇳
#ISRO #NASA

[Andrej Karpathy @karpathy]

So so so cool. Llama 1B batch one inference in one single CUDA kernel, deleting synchronization boundaries imposed by breaking the computation into a series of kernels called in sequence. The *optimal* orchestration of compute and memory is only achievable in this way.

[Andrew Ng @AndrewYNg]

Agentic Document Extraction just got much faster! From previous 135sec median processing time down to 8sec. Extracts not just text but diagrams, charts, and form fields from PDFs to give LLM-ready output. Please see the video for details and some application ideas.

[Satyabrata pal @TheCodingProjec]

@rasbt wish you a speed recovery @rasbt

[Dr. S. Jaishankar @DrSJaishankar]

The world must show zero tolerance for terrorism.

#OperationSindoor

[elvis @omarsar0]

BREAKING: OpenAI introduces new o-series models

o3 and o4-mini

OpenAI claims that these are models that can produce novel and useful ideas.

Here is all you need to know:

[Satyabrata pal @TheCodingProjec]

@MeganRisdal @kaggle @GoogleColab Wow! kaggle surprises me everytime.