---
title: "Rounak Show (@ShowRounak) | Hashnode"
description: "Learning Data Science. Read the latest articles by Rounak Show on Hashnode."
url: "https://hashnode.com/@ShowRounak"
---

@ShowRounak

# Rounak Show

@ShowRounak · Remote · Joined February 2023

Learning Data Science

### [Advanced Embeddings from Scratch — Part 3: CLIP (How Text and Image Embeddings Learn a Shared Semantic Space](https://rounakshow.hashnode.dev/clip-how-text-and-image-embeddings-learn-a-shared-semantic-space?utm_source=hashnode&utm_medium=feed)

In Module 1, we understood why we need multimodal embeddings and how two different encoders can learn compatible representations. In Module 2, we opened the image encoder and learned how an image beco

16 min read · [rounakshow.hashnode.dev](https://rounakshow.hashnode.dev/clip-how-text-and-image-embeddings-learn-a-shared-semantic-space?utm_source=hashnode&utm_medium=feed)

[1](https://hashnode.com/posts/clip-how-text-and-image-embeddings-learn-a-shared-semantic-space/6af759e358c56e403661e022) [Nov 13](https://hashnode.com/posts/clip-how-text-and-image-embeddings-learn-a-shared-semantic-space/6af759e358c56e403661e022)

### [Advanced Embeddings from Scratch — Part 2: How Do Images Become Embeddings?](https://rounakshow.hashnode.dev/how-do-images-become-embeddings?utm_source=hashnode&utm_medium=feed)

In Module 1, we learned that multimodal models can represent text and images in a shared embedding space. We also learned how contrastive learning encourages matching text-image pairs to have compatib

18 min read · [rounakshow.hashnode.dev](https://rounakshow.hashnode.dev/how-do-images-become-embeddings?utm_source=hashnode&utm_medium=feed)

[Nov 12](https://hashnode.com/posts/how-do-images-become-embeddings/6af5cb5322ac8da54b0bbf30)

### [Advanced Embeddings from Scratch — Part 1: Why Do We Need Multimodal Embeddings?](https://rounakshow.hashnode.dev/why-do-we-need-multimodal-embeddings?utm_source=hashnode&utm_medium=feed)

In our original Understanding LLM Embeddings from Scratch series, we learned how text becomes vectors, how those vectors learn semantic relationships, and how sentence embedding models represent entir

14 min read · [rounakshow.hashnode.dev](https://rounakshow.hashnode.dev/why-do-we-need-multimodal-embeddings?utm_source=hashnode&utm_medium=feed)

[Nov 12](https://hashnode.com/posts/why-do-we-need-multimodal-embeddings/6af574626c71095c8801b398)

### [Inside Attention: Why Softmax Sits Between QKᵀ and V](https://rounakshow.hashnode.dev/why-do-we-apply-softmax-to-qk-before-multiplying-by-v?utm_source=hashnode&utm_medium=feed)

When I first learned self-attention, the formula looked straightforward: $$\\operatorname{Attention}(Q,K,V)=\\operatorname{softmax}\\left(\\frac{QK^\\top}{\\sqrt{d\_k}}+M\\right)V$$ But one question bothered

6 min read · [rounakshow.hashnode.dev](https://rounakshow.hashnode.dev/why-do-we-apply-softmax-to-qk-before-multiplying-by-v?utm_source=hashnode&utm_medium=feed)

[Nov 9](https://hashnode.com/posts/why-do-we-apply-softmax-to-qk-before-multiplying-by-v/6af1b85942416b89d2ff0a96)

### [Understanding LLM Embeddings from Scratch — Part 8: from an Inference Engineering Perspective](https://rounakshow.hashnode.dev/understanding-llm-embeddings-from-scratch-part-8-from-an-inference-engineering-perspective?utm_source=hashnode&utm_medium=feed)

Hey folks, this is the final module of our Understanding LLM Embeddings from Scratch series. Across the previous seven modules, we've learned how text becomes token IDs, how embeddings are learned, ho

8 min read · [rounakshow.hashnode.dev](https://rounakshow.hashnode.dev/understanding-llm-embeddings-from-scratch-part-8-from-an-inference-engineering-perspective?utm_source=hashnode&utm_medium=feed)

[Nov 4](https://hashnode.com/posts/understanding-llm-embeddings-from-scratch-part-8-from-an-inference-engineering-perspective/6aeb4ffc80dd863728d7ba95)
