Just the code

If you'd like to repeat or build upon some of my AI-adjacent experiments, you're welcome to grab the code from GitHub. Full disclosure: I do not have formal development training and create code with the help of all-too-fallible AI co-pilots. Check code carefully before you run it in your environment.

Blog generator GPT-4

This Python script is intended to combat model "laziness" by prompting OpenAI's GPT-4 to first generate an outline and then feed it the outline one section at a time for copywriting. The system prompt includes information on the blogger's preferred audience and voice as well as writing samples and style guidance drawn from a text file. GitHub repo.

Blog generator Claude

Version 1 of this Python script prompts Claude to first generate a blog outline and then a complete blog post based on the outline. The system prompt includes information on the blogger's preferred audience and voice as well as writing samples and style guidance drawn from a text file. GitHub repo.

Version 2 prompts Claude to generate posts by feeding it one section of the outline at a time. It also saves the outline and blog copy in a .csv file with columns named Section Name, Blog Content, and Outline Content and supports the same detailed system prompt described above. GitHub repo.

Edit captions JSON format

This Python script reads image captions from a CSV file, shortens them using OpenAI's API, and then saves them in a JSON file. This could be particularly useful if you need image captions equal to or less than a certain character count for accessibility, social media, or training an image model. GitHub repo.

Image captions

This Python script uses OpenAI's GPT-4-Turbo model to generate image captions and then store them alongside the corresponding image file names into a .csv file. It's useful if you need to generate numerous captions for updating alt tags on your website, training machine learning models, etc. GitHub repo.

Image shrinker

This Python script will automatically resize images and maintain their aspect ratio. It's useful for batch processing social graphics and preparing image files to fine-tune or train image generation models. GitHub repo.

Text-to-speech + copy generation

This MacOS desktop application lets you generate an audio script using OpenAI's GPT4-Turbo model and record spoken audio using AI voices from OpenAI and ElevenLabs. It provides an intuitive interface for testing a wide variety of AI voices and generating audio content. GitHub repo.