Post

https://twitter.com/hasantoxr/status/1724767521579139483

More from @hasantoxr

Hasan Toor

@hasantoxr

Dec 30

R.I.P generic prompting.

Context engineering is the new king.

OpenAI, Anthropic, and Google engineers don't write prompts like everyone else. They engineer context.

Here are 10 ways to use context in your prompts to get pro-level output from every LLM out there:

1/ PERSONA + EXPERTISE CONTEXT (For any task)

LLMs don't just need instructions. They need to "become" someone. When you give expertise context, the model activates completely different reasoning patterns.

A "senior developer" prompt produces code that's fundamentally different from a generic one.

Prompt:

"You are a [specific role] with [X years] experience at [top company/institution]. Your expertise includes [3-4 specific skills]. You're known for [quality that matters for this task].

Your communication style is [direct/analytical/creative].

Task: [your actual request]"

2/ REFERENCE CONTEXT (For websites & content)

All of us know this that LLMs hallucinate when they guess what you want.

When you show them exactly what "good" looks like, they stop guessing. Reference context transforms vague requests into precise execution.

This is how agencies get consistent brand voice across hundreds of outputs.

Prompt:

"REFERENCE EXAMPLES:
[Paste 2-3 examples of the style/format you want]

WHAT MAKES THESE WORK:
- [Pattern 1 you noticed]
- [Pattern 2 you noticed]
- [Pattern 3 you noticed]

NOW CREATE:
[Your specific request] following the exact patterns above. "

Read 11 tweets

Hasan Toor

@hasantoxr

Dec 29

BREAKING: I stopped wasting hours reading textbooks cover to cover.

NotebookLM now teaches me directly from PDFs and notes.

Here are 9 prompts that turned documents into lessons:

1. Big Picture Breakdown

Prompt:
“I uploaded this PDF. Give me a high-level overview of the entire document, broken into key themes and concepts, as if you’re introducing it to someone seeing it for the first time.”

2. Teach Me Like a Student

Prompt:
“Teach the content of this document step by step, starting from the basics and gradually increasing difficulty. Assume I’m learning this subject for the first time.”

Read 11 tweets

Hasan Toor

@hasantoxr

Dec 27

🚨 HOT TAKE: Google Research just dropped the textbook killer.

Its called "Learn Your Way" and it uses LearnLM to transform any PDF into 5 personalized learning formats. Students using it scored 78% vs 67% on retention tests.

The new education revolution is here.

The issue with textbooks is that they're the same for everyone and can be really boring.

Google's idea is genius: AI checks out what you like (like sports, music, or food) and knows your grade level, then changes the examples to fit what you actually care about.

Physics starts to make sense, and history feels more connected to you.

But it doesn't stop at personalization.

The system generates 5 completely different ways to learn the same content:

- Interactive text with embedded questions
- Audio lessons with AI teacher conversations
- Narrated slides with fill-in-the-blanks
- Mind maps you can zoom in/out
- Section quizzes

Read 8 tweets

Hasan Toor

@hasantoxr

Dec 25

I used to spend 20 hours a week on content. Now I spend 45 minutes.

The difference? I stopped writing prompts and started building a "Studio Engine."

Here are the 12 prompts I used to scale our output by 600% while increasing our engagement rate by 22%.

This is the most valuable system I’ve built all year. 🧵

1/ The “Voice Injection” Prompt

Gets the model to fully absorb your writing style.

“Here are 5 samples of my writing. Extract my tone, pacing, sentence structure, and emotional signatures. Confirm when my ‘voice profile’ is ready.”

This sets the foundation.

2/ The “Idea Engine” Prompt

Infinite hooks, titles, and angles.

“Generate 20 content ideas that could go viral in my niche. Mix curiosity, tension, and contrarian angles. No clichés.”

It’s like having a strategy intern.

Read 14 tweets

Hasan Toor

@hasantoxr

Dec 23

Prompt engineering is dead.

Anthropic just published their internal playbook on what actually matters: XML-structured prompting.

Only 2% of users know this exists.

Here's what changed:

Anthropic's engineers built Claude to understand XML tags.

Not as code.

As cognitive containers.

Each tag tells Claude: "This is a separate thinking space."

It's like giving the model a filing system.

The difference is brutal.

Standard prompt: "Write a product description for running shoes considering comfort, durability, and style."

Claude gives you generic output.

Tagged prompt:

Write a product description
running shoes

- comfort
- durability
- style

Output quality jumps 40%.

Read 16 tweets

Hasan Toor

@hasantoxr

Dec 20

Top engineers at OpenAI, Anthropic, and Google don't prompt like you do.

They use 10 techniques that turn mediocre outputs into production-grade results.

I spent 2 weeks reverse-engineering their methods.

Here's what actually works (steal the prompts + techniques) 👇

Technique 1: Constraint-Based Prompting

Most prompts are too open-ended. Engineers add hard constraints that force the model into a narrower solution space, eliminating 80% of bad outputs before they happen.

Template:

Generate [output] with these non-negotiable constraints:
- Must include: [requirement 1], [requirement 2]
- Must avoid: [restriction 1], [restriction 2]
- Format: [exact structure]
- Length: [specific range]

Example:

Generate a product description for wireless headphones with these constraints:
- Must include: battery life in hours, noise cancellation rating, weight
- Must avoid: marketing fluff, comparisons to competitors, subjective claims
- Format: 3 bullet points followed by 1 sentence summary
- Length: 50-75 words total

Technique 2: Multi-Shot with Failure Cases

Everyone uses examples. Engineers show the model what NOT to do. This creates boundaries that few-shot alone can't establish.

Template:

Task: [what you want]

Good example:
[correct output]

Bad example:
[incorrect output]
Reason it fails: [specific explanation]

Now do this: [your actual request]

Example:

Task: Write a technical explanation of API rate limiting

Good example:
"Rate limiting restricts clients to 100 requests per minute by tracking request timestamps in Redis. When exceeded, the server returns 429 status."

Bad example:
"Rate limiting is when you limit the rate of something to make sure nobody uses too much."
Reason it fails: Too vague, no technical specifics, doesn't explain implementation

Now explain database indexing.

Read 11 tweets

Share this page!

Enter URL or ID to Unroll

Hasan Toor

Try unrolling a thread yourself!

More from @hasantoxr

Hasan Toor

Hasan Toor

Hasan Toor

Hasan Toor

Hasan Toor

Hasan Toor

Did Thread Reader help you today?

Don't want to be a Premium member but still want to support us?

Send Email!