As a traditional software developer, have you been watching all the AI advancements and wondering when you'll have enough time to acquire the background in Machine Learning and Data Science they say you need before you can enter the field?
What if I told you that there's an alternate pathway available for people who are already have some programming skills? And that you can complete that pathway in under 8 hours.
The key is to stop reading blog posts and watching YouTube videos and instead start building an actual AI-powered application. I'm willing to bet you acquired your current dev chops by building hands-on apps. You'll make the jump to AI development by hands-on learning not more reading.
But, you have to get started with the right tools and technologies or else you run the risk of spinning your wheels in the fringes of AI without learning the skills you're going to need.
The Power of Focus
AI is a huge field. You could spend months in Support Vector Machines and Decision Trees and be no further in your journey than when you started.
So let's ignore the fringe fields of AI that you won't be using. There will be plenty of time to learn those later.
What does your boss think when he hears the word "AI"?
ChatGPT.
What example does your product owner have in mind when she asks your team to build an AI bot for your product offerings?
ChatGPT.
Is it obvious yet? The most effective way to make the jump to immediately useful AI development is with LLMs.
But GPT-4 cost about $100 million of compute power to train. How can you even use LLMs as a regular software developer?
Here's the Good News
The vast majority of AI bots you'll ever need can be built using pre-trained open-source models. And there are a lot of open source models being released these days.
In fact, there are only two approaches to LLMs you need to know to become the top AI expert on most traditional dev teams in 2024:
- Retrieval-Augmented Generation (RAG) where you feed data from a database back to the user as though the model had been trained on it.
- Fine-Tuning a Model where you adjust a pre-existing model’s weights with your own data.
Why This Course is for You
I've trimmed off the fat so that you can level up faster than you ever dreamed. In fact, the entire course can be completed in less than eight hours.
You could literally get off work on Friday as a traditional software developer and walk back into the office on Monday having added an AI skillset that makes you the top AI expert on your team.
What You’ll Learn
In this course, I’ll teach you how to build your own custom chatbot on top of an open-source, pre-trained LLM.
Here’s what you can expect:
Module 1: From Zero to Chatting with an LLM
⛏️: Get set up with the 3 key AI tools you need to begin building your chatbot. Then get an LLM-powered chatbot working on your machine. You will be chatting with it by the end of the first module.
Module 2: Getting Smarter by Using Outside Sources
⛏️: Learn the secrets behind teaching your LLM to behave as a custom chatbot instead of a general language model. Then hook up your LLM with your custom data, so that it can give conversational answers that appear to have come from the model via the process of retrieval augmented generation (RAG).
Module 3: Bake Data into your LLM
⛏️: Discover how to bake your custom data into an already pre-trained LLM model via the process of fine-tuning. This means you can have a truly custom LLM but avoid the expense of having to train one from scratch.
Note that Module 3 is currently in development and is planned for an early September release. By purchasing the course at the founding member price you will automatically have access to module 3 when it releases.
Project: The Cat Café
⛏️: Throughout the course, to get hands on experience, we will be building a conversational pet adoption bot for a local cat café. Every day, new cats arrive and are added to the database via adoption poster PDFs. By the time you're done your, your bot will be able to answer queries such as, “I’m a 60 year old reclusive author looking for a feline companion to help me binge watch Yellowstone.”