**Coding with AI: Harnessing the Power of Large Language Models**
---
**Today's Journey**
- Principles
- Live demos
- Putting it into action
- Q&A
---
**Two principles**
- Making things with LLMs is an iterative process
- To get the best results think: "open book test"
---
**Coding with AI should feel like a conversation**
- Iterative process
- You basically have to have a conversation with it to get the desired output
- https://chat.openai.com/share/5fce83eb-6f72-4e91-84a1-b2aeb61ecc82
---
**ChatGPT is not a knowledge database, it's a resoning engine**
A detective is really good at solving puzzles and figuring out clues, but they need to have the right information to solve the case.
A reasoning engine is like that detective. It's really good at thinking things through and making guesses, but it needs the right information to be even better at solving problems.
---
- Data cuts off: September 2021
- This means it won't have access to latest docs and API's
---
**Giving AI Open Book Tests**
- Provide ample context
- Give it a way to answer your question as well as the question
- https://every.to/chain-of-thought/gpt-4-is-a-reasoning-engine
---
**Exercise 1: ChatGPT & LinkedIn**
- Extracting profile details
- Using the system prompt for accuracy
---
**Exercise 2: Transforming Websites into Chatbots**
- Choose a website
- Scrape its data
- Put it into an LLM
- Have a conversation with it
---
**Exercise 3: Making a GPT-4 command line bot**
- We're going to use Replit and python to do this
---
**Wrapping Up**
- Questions?
- Thank you for participating!
- Resources & follow-up sessions
{"title":"Coding with AI","description":"Of course! Here’s the content for your slides in markdown format:","contributors":"[{\"id\":\"8a2b9d3f-73af-48e2-bb38-13888be4577f\",\"add\":2656,\"del\":1027}]"}