Hemant Pandit

INDIAN

[Alex Xu @alexxubyte]

OAuth 2.0 Flows

[Nam Dinh @namd1nh]

🧠 Still confused between AI, ML, DL, LLMs, GenAI & Agentic AI?

You're not alone. Here's a crash course 🧵

Let’s break it down — plain & simple ↓

[Alex Xu @alexxubyte]

Session, Cookie, JWT, Token, SSO, and OAuth 2.0 Explained in One Diagram

[Alex Xu @alexxubyte]

What is an AI Agent?

[Alex Xu @alexxubyte]

RAG vs Agentic RAG

[Akshay 🚀 @akshay_pachaar]

RAG vs. Agentic RAG, clearly explained:

[Alex Northstar @NorthstarBrain]

FINALLY! OpenAI made a guide that literally explains WHEN to use WHAT AI model

What you need to know + example prompts to bookmark:

[Bindu Reddy @bindureddy]

Pretty good explanation of agentic design patterns.

We use all of them in DeepAgent

[ℏεsam @Hesamation]

Transformers in Excel must be the most cracked thing I've seen.

this has everything
• Positional Encoding
• Self-Attention
• Cross-Attention
• Multi-head Attention
• Skip Connection
• LayerNorm
• ReLU Activation
• Feed Forward
• Softmax

[munen @munen5647]

Nobody can explain Transformers and Self-Attention like professor Bryce...
He's one of the Hidden gems of YouTube, his all videos are packed with knowledge and he teach with enthusiasm and dedication.
Below 👇🏻 link in comments.

[ₕₐₘₚₜₒₙ @hamptonism]

Artificial Intelligence Road Map:

[Akshay 🚀 @akshay_pachaar]

Let's compare DeepSeek-R1 and OpenAI-o1 using RAG:

[Avi Chawla @_avichawla]

5 LLM fine-tuning techniques, clearly explained (with visuals):

[Maxime Labonne @maximelabonne]

🆕 LLM Course 2025 edition!

I updated the LLM Scientist roadmap and added a ton of new information and references.

It covers training, datasets, evaluation, quantization, and new trends like test-time compute scaling.

💻 LLM Course: https://t.co/q5XtSIC7nL

[Nagesh Naik @nageshvnaik]

Goosebumps 🥺🫶🏻

[Aina Sanghi @ainasanghi]

Git commands cheatsheet

[Akshay 🚀 @akshay_pachaar]

Let's learn how to evaluate a RAG application (part 1):

1/n

[amriting @saamaanyafreaky]

Massive dump of 50+ unsorted mobile wallpapers

🧵

[Troll Football @TrollFootball]

Yes, I am an Indian and a proud Indian.

Vande Mataram

[Akshay 🚀 @akshay_pachaar]

Great overview of how to build your own Document Chat RAG!

Here's what you'll need:

1️⃣ A knowledge base

A collection of relevant and up-to-date information that serves as a foundation for RAG. It can be a database, a set of documents, or a combination of both. In this case it would be the collection of your documents.

2️⃣ Chunking strategy

Chunking is the process of breaking down a large input text into smaller pieces. This ensures that the text fits the input size of the embedding model and improves retrieval efficiency.

Two hyperparameters that you need to take care of are the `chunk_size` & `chunk_overlap`.

3️⃣ Embeddings strategy

The chunked data is converted into embeddings (vector representations), here you need to decide what embedding model to use.

4️⃣ Document retriever

A document retriever is a crucial component that performs the task of searching and retrieving semantically similar documents or chunks based on a given query.

5️⃣ Context generation

Now you need to create a context based on the retrieved, here's what we usually do:

- keep top-k chunks
- put a similarity cutoff
- filtering based in metadata
- reranking chunks using a reranker model

6️⃣ Prompt template

Use a custom prompt template to guide the response from LLM and ensure it aligns with the context provided.

7️⃣ Choose Your LLM

Next, select the LLM you want to use. It will generate the final response based on the prompt you created in the last step.

8️⃣ Evaluating the responses

Now considering all the hyperparameters that we have, we tune them by trying different combinations, here are some of the metrics for evaluating the final response:

- BLEU score
- METEOR score
- BERT score
- ROGUE score

Finally you select the best model & deploy this system into production!

Credits: @abacusai