AI Engineering Day 2: Reasoning vs Non-reasoning models and adding custom parameters to API call

Video Description

In this video, we dive into difference between reasoning and non-reasoning models and fiddling with model parameters, first using OpenAI's chat playground and then within our AI app in Cursor Links: - uv tutorial: https://youtu.be/W4pmn5zD7b4?si=spGfiSaksJgTjojT - Code at the start of this video : https://github.com/MathNMachines89/ai_engineering_day1 - Code generated at the end of this video: https://github.com/MathNMachines89/ai_engineering_day2 - OpenAI chat playground: https://platform.openai.com/chat/edit?models=gpt-5 - OpenAI model overiew: https://platform.openai.com/docs/models - OpenAI model comparative pricing: https://platform.openai.com/docs/pricing - Python dictionary unpacking operator: https://pieriantraining.com/python-unpack-dictionary-a-comprehensive-guide/ Python Learning Resources: - https://www.youtube.com/watch?v=eWRfhZUzrAc (freecodecamp) - https://www.w3schools.com/python/ Timecodes 00:00 - Intro 00:27 - OpenAI's Completions API call 01:28 - OpenAI's chat playground and custom parameters 02:39 - Reasoning vs Non-Reasoning models 06:46 - Nano vs Mini vs Normal model versions 07:55 - What is a System Message? 09:57 - What is a Developer Message? 11:22 - OpenAI Models Overview 12:47 - Constructing prompt to add custom parameters to our API call 14:52 - Reviewing AI's Pull Request 17:40 - Running AI app on command line 18:45 - Asking AI to resolve an error 20:16 - Why System/Developer messages are useful? 21:29 - Cursor's inline edit feature Featured Playlist: https://www.youtube.com/playlist?list=PLLfyHtsENKgayfsI5h3A2qm4iziibIZdQ