He grew up poor, made millions with his companies and then lost all of his fortune. He went literally bankrupt.

This happened not just once, but twice.

Now he's worth 100+ millions. So he definitely knows about wealth, maybe more than anyone else.

Few months ago, Scott Galloway shared the 5 rules for building wealth in his book, the algebra of wealth. And these rules are a must-read for anyone who want to be financially successful.

Here’s the breakdown:

Rule #1: Define a goal besides money

Money is should not be the goal, but the mean to an end.

If you want to make money for the sake of being richer you will never have enough.

Instead, you need to understand what is your inner motivation. In my case, I want to become wealthy so that I can:

• Provide a better life for the people I care about

• Improve my quality of life (bigger bank account = better healthcare, better food, better life)

• Work because I want to, not because I have to

Ask yourself what are the reasons why you want to gain financial security. Remember them in your journey.

Rule #2: Don't let your impulses control you. Control them instead

We all have emotions, and sometimes it is hard to remain control them. But what does control mean here?

It doesn't mean repressing your emotions or ignoring them. It means acknowledging their existence without letting them take the steering wheel.

Getting angry is ok. Yelling at others because you are angry is not.

Rule #3: Exercise

It takes people a long time to learn this one

This might seem like a very generic advice. But it has strong scientific foundations. There is a direct link between successfull people and exercise.

Exercising improves the state of your body, boost your energy and allows you to let out emotions in a healthy way (see rule #2).

There's a reason if 90%+ of millionaires train at least 3 times per week.

Rule #4: Make rich friends

Especially if you were not born rich.

Why? Not because they are better human beings all around, but becuase they are some steps ahead of you in the game.

And you can learn something form them. Not only how do they invest or what trend they think is worth looking at.

The #1 most important thing is the network. The closer you are to money, the more interconnected you are to money, the more likely you are to be part of it.

Rule #5: Talk about money

We are often taught from a young age that talking money is not ok. It's a taboo.

You shouldn't ask someone their salary.

You shouldn't ask them about their investments.

Guess who doesn't talk about money? Poor people

Guess who does? Rich people

Stop thinking about wealth and money as a taboo. Start asking your friends how much they make. Where do they invest? How much do they invest.

However, don't make the mistake of believing everything. Losses and mistakes are often not talked about (because they are painful experiences, better to talk about wins)

Retrieval augmented generation (RAG) is a technique to combine information retrieval with an LLM. It allows the foundational model to use a set of documents as a knowledge base.

Think about how we use Notion or Obsidian as a second brain. It's a good analogy for RAG. It lets our model be up-to-date with the latest knowledge and reduces hallucinations

(Cheap) Superpowers for your LLM

RAG often solves the some pain points that make building LLMs application hard:

• Reduced hallucination by providing context to your model

• Easy updates: It allows using recent data without long and expensive fine-tuning cycles

• Factual transparency: your model can accurately reference relevant documents

Now that we have seen an overview of use cases, let's see what is RAG and how it works.

What is RAG?

The idea behind RAG is very simple: whenever we prompt the LLM, we look at set of documents (knowledge base) and find the items most similar to our query. Then we pass the prompt and the context to the model.

On a high level, the RAG architecture uses is based on 3 things:

• Language model

• Knowledge base

• Similarity measure

Popular RAG architectures

Let's see what are the 3 most popular architectures for Retrieval-Augmented generation systems.

Basic RAG: this architecture was introduced in the paper Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. It combines a retrieval componenet with a sequence to sequence model.

How it works:

• The input query is encoded and used to retrieve relevant documents from an index

• The selected documents are concatenated to the input query

• The query + documents are fed to a language model, which generates the output

Fusion-in-Decoder: this approach comes from the paper Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering, and it extends the Basic RAG strategy. How it works:

• (same as Basic RAG) The input query is encoded and used to retrieve relevant documents from an index

• Each retrieved document is processed separately by an encoder

• Attention is used on encoded queries + documents simultaneously

Retrieval-Enhanced Transformer: from the paper Improving language models by retrieving from trillions of tokens. This approach incorporates the retrieval into the transformer architecture.

How it works:

• At each layer of the Transformer, retrieval is performed based on the current hidden states

• The retrieved information is then incorporated into the next layer

Choosing a system

Each RAG system has pros and cons, and may be more suitable for different tasks or constraints.

Keeping this in mind, there isn't a one-size-fit-all RAG system, as each has its own tradeoffs. The best RAG architecture to use is the one that best aligns with your functional and non functional requirements.

We are surrounded by Large Language Models (LLMs) like ChatGPT, and their capabilities seems almost magic... when someone else is using them.

But often when it's our turn to tap into this powerful technology we get way too basic answers, sometimes they are so dumb it's annoying.

What can we do about this?

Ask things the right way

Imagine entering a room where there is a meeting going on. As soon as you get in, people start asking you questions about what they are discussing. You would be caught off guard and probably your answers will not be great.

This is what's happening to chatGPT most of times.

To get good answers we need to give it the context before asking the questions. This is called prompting.

Prompting can make a huge difference in the quality of the answers, since the AI

• Will understand what we really need

• Will use the right information to help us

• Will know how much detail we want in the answer

What is a good prompt?

• Give background info before asking questions

• Be clear about what you want

• Include important details about your situation

• Ask for step-by-step plans and ways to measure success (if applicable)

• Request examples when you need them

• Try different ways of asking to see what works best

Here's an example where we compare a badd prompt and a good one: we are asking ChatGPT how to grow our company social media presence. However we are asking it in different ways. Go to chatgpt website and try it yourself:

Bad prompt:

How do I grow my company social media presence?

Good prompt:

Provide a detailed and actionable strategy for growing the social media presence of my company.

My company is a [type of business] that primarily operates in the [industry or niche]. We currently have [X] followers on [list of social media platforms, e.g., Instagram, Twitter, LinkedIn, etc.], but we aim to significantly increase our online presence. Our target audience includes [specific demographics or customer profiles]. The strategy should take into account our brand and current trends

The goal is to receive a comprehensive plan that includes specific tactics, timelines, and measurable KPI (key performance indicators). The strategy should outline both organic and paid growth methods. Additionally, provide recommendations for tools or platforms that could enhance our social media management and analytics.

If possible, include examples or case studies that demonstrate successful social media growth in a similar industry.

This is game changing.