Post

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

More from @hasantoxr

Hasan Toor

@hasantoxr

Dec 2

Holy shit… I just found out why OpenAI, Anthropic, and Google engineers never worry about prompts.

They use context stacks. Context engineering is the real meta.

It’s what separates AI users from AI builders.

Here's how to write prompts to get best results from LLMs:

Prompt engineering was a hack for the early days of AI like learning to talk to a foreigner using short phrases and keywords.

But today’s models don’t just understand instructions. They understand environments.

Your job isn’t to “prompt” the model.

It’s to architect its context.

Think of context as a digital environment you build around the model before it ever starts generating.

You define:

• Who it should “be” (role/persona)
• What it’s trying to achieve (goals)
• How it should communicate (tone/style)
• What to reference (examples, data, past work)

This is what drives consistent, on-brand, high-quality output.

Read 10 tweets

Hasan Toor

@hasantoxr

Nov 30

Gemini 3.0 Pro just killed consulting as we know it.

Here are the 3 prompts I use to get McKinsey-grade answers instantly 👇

Let me tell you what McKinsey consultants actually do:

1. Analyze industry trends and competitive dynamics
2. Benchmark companies and products
3. Identify strategic risks and opportunities
4. Package it all in fancy slides and charge 6 figures

But guess what?

AI can now do 90% of that instantly.

Let me show you how:

We use these 3 mega prompts for different tasks:

1/ The Consultant Framework

Prompt: "You are a world-class strategy consultant trained by McKinsey, BCG, and Bain. Act as if you were hired to provide a $300,000 strategic analysis for a client in the [INDUSTRY] sector.

Here is your mission:

1. Analyze the current state of the [INDUSTRY] market.
2. Identify key trends, emerging threats, and disruptive innovations.
3. Map out the top 3-5 competitors and benchmark their business models, strengths, weaknesses, pricing, distribution, and brand positioning.
4. Use frameworks like SWOT, Porter’s Five Forces, and strategic value chain analysis to assess risks and opportunities.
5. Provide a one-page strategic brief with actionable insights and recommendations for a hypothetical company entering or growing in this space.

Output everything in concise bullet points or tables. Make it structured and ready to paste into slides. Think like a McKinsey partner preparing for a C-suite meeting.

Industry: [INSERT INDUSTRY OR MARKET HERE]"

Read 7 tweets

Hasan Toor

@hasantoxr

Nov 29

A brand just developed a comprehensive content plan, built a complete calendar, and published automatically with a single AI tool in one hour.

Pippit AI enables you to generate high-converting videos, images, and social content seamlessly.

Here’s how it works:

This is the world’s FIRST Marketing Agent does EVERYTHING powered by Vibe Marketing

Create content plans from one sentence, including topics, scripts, hooks, and hashtags.

Make videos and images quickly, matching each platform's style.

Set up a content calendar with daily posts. automatically publish to TikTok, Instagram, and Facebook without manual uploads.

pippit.ai/thoughtform-on…

One message to assets and full content plan
Enter your prompt -> Add your launch details

-> Auto-analyze competitors & platform trends -> Auto-generate assets & a publishing calendar
-> Publish

Read 6 tweets

Hasan Toor

@hasantoxr

Nov 27

🚨 This project just made most AI agents look outdated.

It’s called Better Agents and it supercharges your coding assistant (Kilocode, Claude Code, Cursor, etc), making it an expert in any agent framework you choose (Agno, Mastra, etc) and all their best practices.

This is the future of autonomous AI.

Here’s how it works 👇

Every AI agent demo looks amazing until production hits.

Then reality:
— Agents hallucinate in edge cases
— No version control for prompts
— Zero test coverage
— Debugging = prayer

Better Agents isn't another framework. It's the testing layer everyone forgot to build.

Link: github.com/langwatch/bett…

Better Agents wraps ANY framework (Agno, Mastra, LangGraph) with:
→ Scenario testing
→ Prompt versioning
→ Auto instrumentation
→ Evaluation notebooks

Think pytest for agents, but actually useful.

Read 11 tweets

Hasan Toor

@hasantoxr

Nov 25

This open-source project just solved the biggest problem with AI agents that nobody talks about. It's called Acontext and it makes your agents actually LEARN from their mistakes.

While everyone's building dumb agents that repeat the same errors 1000x, this changes everything.

Here's how it works (in plain English):↓

Acontext built a complete learning system for agents:

— Store: Persistent context & artifacts
— Observe: Track tasks and user feedback
— Learn: Extract SOPs into long-term memory

When your agent completes a complex task, Acontext:
→ Extracts the exact steps taken
→ Identifies tool-calling patterns
→ Creates reusable "skill blocks"
→ Stores them in a Notion-like Space

GitHub: github.com/memodb-io/Acon…

Here's the genius architecture:

User ↔ Your Agent ↔ Session (stores messages)
↓
Task Extraction
↓
Task Completion
↓
Skill Learning
↓
Skill Blocks (stored)
↓
Search & Reuse

Read 9 tweets

Hasan Toor

@hasantoxr

Nov 12

Holy shit... Someone just solved the biggest AI privacy problem nobody talks about.

It's called Hyperlink which is a fully offline AI that searches ALL your documents without sending a single byte to the cloud.

No OpenAI. No Google. No data leaks.

Here's how:↓

Hyperlink flipped the entire model.

Instead of uploading files to AI, they brought AI to your files.

- 100% on-device processing
- Zero cloud dependencies
- Your data never leaves your machine

It's like having GPT-4 intelligence that runs entirely offline.

Here's what's insane:

It indexes EVERYTHING on your computer in real-time:

- PDFs, Word docs, PowerPoints
- Screenshots and images (OCR built-in)
- Meeting notes, legal docs, personal journals

Then lets you search it all with natural language.