sabarinathan

3K posts

sabarinathan banner
sabarinathan

sabarinathan

@sabarinathan_7

Researcher in deep learning, computer vision and machine learning algorithms. I tweet about deep learning and machine learning. Some funny stuff also

Shibuya-ku, Tokyo Katılım Nisan 2013
744 Takip Edilen390 Takipçiler
sabarinathan retweetledi
Allen Braden
Allen Braden@allen_explains·
This 2-hour Stanford lecture breaks down how models like ChatGPT and Claude are actually built, clearer than what many people in top AI roles ever get exposed to. Save this and set aside two hours today. It might end up being the most valuable thing you learn all week.
English
162
4.5K
27.6K
3.4M
sabarinathan
sabarinathan@sabarinathan_7·
🚀 Interesting paper: Attention Residuals (AttnRes) A small but powerful idea to improve Transformers 👇
English
4
0
0
15
sabarinathan
sabarinathan@sabarinathan_7·
Why this matters: Better feature usage More stable training Stronger reasoning performance My takeaway: We already use attention across tokens… Now extending it across layers makes a lot of sense
English
0
0
0
6
sabarinathan
sabarinathan@sabarinathan_7·
This paper proposes: 👉 Use attention across layers Instead of equal addition, the model learns which layers to focus on
English
0
0
0
5
sabarinathan
sabarinathan@sabarinathan_7·
Today’s Transformers use residual connections → All previous layer outputs are added equally But… not all layers are equally useful 🤔
English
0
0
0
9
sabarinathan
sabarinathan@sabarinathan_7·
🚨 Interesting shift in the AI Agents ecosystem Recently I came across an article mentioning that Perplexity is moving away from MCP (Model Context Protocol) and relying more on direct APIs for its agent workflows.
English
10
0
3
32
sabarinathan
sabarinathan@sabarinathan_7·
That said, MCP (Model Context Protocol) is still a powerful idea. It standardizes how AI agents interact with tools and services. But real-world systems often prioritize simplicity and performance.
English
0
0
0
12
sabarinathan
sabarinathan@sabarinathan_7·
Cost control becomes easier too. You can measure every call, implement caching strategies, and optimize usage without additional tool-layer overhead.
English
0
0
0
8
sabarinathan
sabarinathan@sabarinathan_7·
Direct APIs also give predictable performance. You control: • Payload size • Serialization • Request patterns Which helps avoid unpredictable agent behavior.
English
0
0
0
9
sabarinathan
sabarinathan@sabarinathan_7·
Another advantage: observability. With direct APIs you can easily use: • HTTP logs • Metrics • Tracing • Retry strategies This makes production debugging much easier.
English
0
0
0
11
sabarinathan
sabarinathan@sabarinathan_7·
Direct APIs also improve security. Instead of routing everything through a tool layer, systems can use direct authentication with the service, making access control easier to manage.
English
0
0
0
12
sabarinathan
sabarinathan@sabarinathan_7·
Why direct APIs are often preferred in production AI systems: • Lower token usage • Less tool description overhead • Smaller context size This helps reduce LLM cost and latency
English
0
0
0
22
sabarinathan
sabarinathan@sabarinathan_7·
Building production AI systems teaches an important lesson: Sometimes direct APIs are more practical than complex agent tooling. Here’s why 👇
English
0
0
0
17
sabarinathan retweetledi
Andrej Karpathy
Andrej Karpathy@karpathy·
I spent more test time compute and realized that my micrograd can be dramatically simplified even further. You just return local gradients for each op and get backward() to do the multiply (chaining) with global gradient from loss. So each op just expresses the bare fundamentals of what it needs to: the forward computation and the backward gradients for it. Huge savings from 243 lines of code to just 200 (~18%). Also, the code now fits even more beautifully to 3 columns and happens to break just right: Column 1: Dataset, Tokenizer, Autograd Column 2: GPT model Column 3: Training, Inference Ok now surely we are done.
Andrej Karpathy tweet media
English
90
175
2.6K
262.9K
Amazon Help
Amazon Help@AmazonHelp·
@sabarinathan_7 We're sorry to know about your issue with the account. We would like to escalate this for you. Please contact us via DM using the 'Send us a Private Message' option below, without sharing your order or account details, and we will do our best to assist. -Hafeez twitter.com/messages/compo…
English
1
0
0
34
sabarinathan
sabarinathan@sabarinathan_7·
@AmazonHelp @amazonIN @AmitAgarwal @ajassy My order 405-4873672-8028367 was canceled and my account was restricted to digital purchases only, citing “multiple returns/refunds.” I’m requesting clarification, as this does not seem accurate. I’m an active Amazon Prime member.
sabarinathan tweet media
English
4
0
1
51