Day four was more string work. It was focused on manipulating strings and doing some built in methods to build some analytics.

Challenge:

You’re processing user-submitted restaurant reviews. Real user input is messy — inconsistent casing, extra spaces, typos in structure. Write a program that:

1. Starts with this list of raw reviews exactly as given:

raw_reviews = [
    "  great burger, loved it! ",
    "FRIES WERE COLD. bad experience. ",
    "  the milkshake was amazing. will return.",
    "salad was okay. nothing special.   ",
    "BROWNIE IS A MUST TRY!!  "
]    

2. Writes a function clean_review(review) that:

• Strips leading/trailing whitespace
• Converts to sentence case (first letter capitalized, rest lowercase — look up which method does this in one step)
• Returns the cleaned string

3. Uses a list comprehension to apply clean_review to every review, storing results in cleaned_reviews
4. Writes a function analyze_reviews(reviews) that returns a dictionary containing:

• "total": count of reviews
• "positive": count of reviews containing words like “great”, “amazing”, “loved”, “must”, “will return”
• "avg_length": average character length of the reviews (rounded to nearest whole number)

5. Prints the cleaned reviews and the analysis dictionary cleanly

Things to figure out: how to check if any of several words appear in a string, and how to calculate an average from a list.

It’s been about a year since I last tried to “vibe code” and I used the normal Claude LLM to do so. The experience was okay, but frustrating. When using the standard chat interface, you have to be very specific about what you’re feeding the chat and make sure that you keep the context window fairly short, because eventually the tool will start hallucinating more and more as you go on.

What is RAG?

If you’ve been trying to keep up with AI, there’s a good chance you’ve come across the term RAG. When I first learned heard the term, I wasn’t exactly sure what it meant form the context of the discussion.

Retrieval-Augmented Generation (RAG) is a technique that enhances the accuracy of large language models (LLMs) by giving them a wider context window and including external (from the LLM’s database) knowledge sources.

Code Tutor AI 🧙🏼‍♂️

An AI-powered interactive CLI coding tutor built with OpenAI Agent SDK.

Code Tutor is built to allow you to learn to code yourself rather than just vibe code your way to success. This AI agent is supportive and uses the Socractic method to walk you, the coding student, through solving the problem on your own with guidance.

🧠 Code tutor is built on the principle that we can use AI to accomplish great things, but we need to protect our ability to think critically and learn.

Agentic AI

Agentic AI is essentially giving AI a specific job that it will execute on a regular basis and giving it the autonomy to do so.

There are five main categories of agentic AI today:

1. Simple Reflex
2. Model-Based Reflex
3. Goal-Based
4. Utility-Based
5. Learning

Simple Reflex Agents

The most basic agentic AI models are reactive and are built to analyze an input from the environment then use condition based logic/rules to make a decision on what to do next. For instance, your thermostat has a sensor that evaluates the temperature of the air (environmental input) and checks it against the conditional rules (if the temp is above 72 degrees F, then turn on cooling until it reaches 72 degrees). Once it evaluates the data against the conditional logic, it performs an action.