How I use LLMs
Andrej Karpathy. Skip to the "ChatGPT memory, custom instructions" chapter at 1:53:29. Karpathy shows what he actually has in his own custom instructions, talks about when he tells ChatGPT to remember a fact versus letting it auto-capture, and notes the trade-offs — closer to the article's "set it up once with intent" framing than most beginner walkthroughs.
AI Expert note
The reasoning about durable preferences is solid; the exact memory behavior may differ by product version, region and workspace setting. Do not store confidential client, employee or legal details as "preferences."
What you should get from this
Understand how a power user separates durable preferences from one-off context.
Watch or know first
Know where custom instructions and memory live in your current ChatGPT settings.
Watch next
Continue through the same learning path with the next curated companion videos.
4 minutesHow To Use Otter AI To Transcribe Audio - Features and Overview
Compare a standalone transcription workflow with integrated meeting assistants before choosing a tool.
Watch next
13 minutesHow to Use Copilot to Automate Meeting Notes in Microsoft Teams!
Understand how integrated assistants can connect email, meetings and tasks into one reviewable work loop.
Watch next
90 minutesThe Ultimate Guide to Writing with AI
Use AI for research, structure and revision while keeping final wording and judgment under human control.
Watch nextRelated videos
Take it further
Hand-picked external courses that go deeper on this topic.
DeepLearning.AIChatGPT Prompt Engineering for Developers
Ninety minutes to a year's worth of intuition. If you've started writing code that calls an LLM — or you're about to — this is the most efficient course online for closing the gap between "playing with ChatGPT" and "shipping a feature that calls an LLM."
Coursera · Vanderbilt UniversityPrompt Engineering for ChatGPT
The academic complement to DeepLearning.AI's short course — same discipline, longer arc, written for people who don't code. Dr. White teaches prompting as a set of reusable patterns (Ask for Input, Outline Expansion, Fact Check List, Menu Actions) rather than tricks. After this you'll prompt LLMs like a designer, not a guesser.