AI Mise en place

Written By Domenic Ashburn

Mise en place = A place for everything. 

A method of cooking clean and efficiently at a high level for the most successful chefs in the world.

How to prepare your space for AI to get quality outputs that are relevant to your palette, enjoy the experience, invite a healthy culture of high performance and have the bandwidth to experiment as new things come out.

INGREDIENTS of AI

The ingredients of AI are actually very simple. 

  • Text

  • Images

  • And sometimes Tables like csv. files which are essentially also just organized text. 

You need to prepare these ingredients for the AI to cook with. So you need to prepare the text and images that are important to your work and life so that AI can use them properly. 

this helps AI to

  • give you answers that are relevant

  • Give you answers that are consistent 

  • Find information fast so you aren’t scrambling to find information and files later.

POTS AND PANS 

How do we provide these ingredients to AI so that it can access them when you’re ready to cook? 

Via: 

  • Google Drive

  • Project Files

  • CustomGPTs

SEASONING 

Custom instructions 

FILE NAMING CONVENTION

Make it easy to access the files by organizing the folders in a naming convention. 

This way at any time you can look up a file and know exactly where it is or you can find it easily by searching for it by keyword. 

https://llmslab.github.io/cookiecutter-ai-flock/tutorials/naming-conventions/

Mr.Gratefuls File Naming: 

YEAR_MM-DD_Department_Title-Version00

2025_07-26_Mindvalley-AI-Summit-Content-v01

Technical Repositories:

<prefix>-<descriptive-name>[-<optional-version>]

  • customer_data_v1.0_20240101_initial.csv

  • sales_report_v2.2_20240305_updated.xlsx

  • image_dataset_v1.0_20240220_raw.json

Prefix: A concise identifier related to the project's domain or main technology.

Descriptive Name: A clear and specific description of the repository's content or purpose.

• Optional Version: A version number, if applicable, to distinguish between different iterations or stages of the project.

Data file names

<dataset_name>_<version>_<creation_date>_<description>.<extension>

  • Dataset Name: A concise identifier for the dataset.

  • Version: Version number or identifier of the dataset.

  • Creation Date: Date when the dataset was created or last modified, in the format YYYYMMDD.

  • Description: A brief, clear description of the dataset or its specific subset.

• Extension: The appropriate file extension (e.g., .csv, .xlsx, .json).

The point of AI is to generate content in the form of text or images or charts. 

The quality of the output the is determined by the quality of the input. 

https://johnbeve.github.io/LLMO-FOIS/submissions/IntegratingOntologiesLLMRAG.pdf


WEEKLY HOUSE KEEPING

  • Add chats to their respective Project folder

  • Delete irrelevant memories

  • Tag folders

DETAILED STEPS

1. Upgrade to teams to protect your data


2. Opt out of sharing data with ChatGPT


3. Input your custom instructions to determine how AI talks to you.

  • Profile > Settings > 

  • Give prompt for getting more straight forward answers and less sycophant-y responses. 

    • Focus on substance over praise. Skip unnecessary compliments or praise that lacks depth. Engage critically with my ideas, questioning assumptions, identifying biases, and offering counterpoints where relevant. Don’t shy away from disagreement when it’s warranted, and ensure that any agreement is grounded in reason and evidence. Don't use emojis. Just speak normally. 


4. Teach AI / Date AI — What does it know about you, or think it does? Validate, correct and provide information. 

  • What are some things you 100% know about me? 

  • What do you assume about me? 

  • What don’t you know about me? 


5. Go to memory and delete anything that is irrelevant. 


6. Connect Google Drive Account

  • Profile > Settings > Connectors > Google Drive (+others)


7.  Set up schedules to get things done automatically. 

  • Simply ask ChatGPT to do a task at a certain time repeatedly

  • Only in models o3 and 4o-mini

  • Settings > Schedules > manage


8. Creating Google Drive folders with naming conventions 

Saves time and makes it easy for AI and You to find files and reference them later.

  • Company HQ > Department > Year > Category > Files

    • File Naming Convention:  

  • YEAR_MM-DD_Department_Title-Version00

9. Making Projects for each department of business

  • Use Naming Convention: Add Timeframe at end

    • FINANCE-JUNE-2025

  • Users can create unlimited projects.

  • Plus: up to 20 files 

  • Pro, Team, Enterprise: 

  • only 10 files can be uploaded at the same time

  • File Size limit: 512MB each

  • Memory remains active in your projects.

  • When you ask a question in a project, ChatGPT prioritizes the project chats and files

  • https://help.openai.com/en/articles/10169521-projects-in-chatgpt


10. Uploading necessary files from Google Drive into Project Chats

New Project >

11. Later creating customGPTs for automations (in AI Mastery)

BLESS YOU. MAY YOU BE HAPPY HEALTHY AND SAFE. WITH GRATITUDE.

DOMENIC

Domenic Ashburn
Previous

Make your product idea real
Next

Streaming