Ruben Hassid Profile picture
Jun 2, 2024 9 tweets 4 min read Read on X
Anthropic just released their first AI educational course on tool use.

7-course links & academic papers.

Here's the link + a summary of each:

#1 → Intro to tool use Image
First, what is tool use?

Tool use allows Claude to extend its capabilities by invoking:
> external tools
> functions

It allows us to write code that can perform specific tasks that Claude wouldn't be able to do.

#2 → Why does it matter? Image
It's a crucial feature that enhances the value & impact of Claude applications.

> Integrate your existing systems
> Extend Claude's capabilities
> Enhance user experience
> Automate complex tasks
> Scale and customize

#3 → How does it work? Image
Tool use involves defining tools for Claude.

1. Define tools with:
> names
> descriptions
> input schemas

2. Provide them to Claude.

3. Claude uses these tools to perform tasks, with results returned and integrated into responses.

#4 → Your first simple tool Image
They are sharing a simple example of tool use:

→ resolving math problems.

Claude struggles at doing complex math, so they provided a calculator tool.

Full example below: Image
#5 → Forcing JSON with tool use

The most interesting way of utilizing the tool use is forcing Claude to respond with structured content like JSON.

→ Simply ask for it.

Define a tool that describes a particular JSON structure. Claude will respond back. Image
#6 → Complete workflow

They shared a diagram of a general overview of the process:

1. You share the prompt
2. Clause uses a tool
3. You extract input & return results
4. Claude uses a tool to answer

The full process is explained below: Image
#7 → Tool use with multiple tools

Use case: build a customer support chatbot for an electronics company

They provide Claude with a suite of tools it can select from.

> get_user
> get_order_by_id
> get_customer_orders
> cancel_order

Here's the link: github.com/anthropics/cou…
Image
I test every major LLM to help me create better content faster - so you do too.

Check @rubenhssd for more.

It's me :) I'm ruben

If you'd like to support me, RTs cover my absurd API costs & SaaS subscriptions.

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Ruben Hassid

Ruben Hassid Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @RubenHssd

Jun 17
BREAKING: Stanford just surveyed 1,500 workers and AI experts about which jobs AI will actually replace and automate.

Turns out, we've been building AI for all the WRONG jobs.

Here's what they discovered:

(hint: the "AI takeover" is happening backwards) Image
→ Workers don't want full automation. They want partnership.

46.1% of tasks got positive automation ratings, but here's the kicker:

Workers prefer "H3 equal partnership" with AI over full replacement. Image
→ Scientists divided AI adoption into 4 zones based on worker desire vs. technical capability.

41% of startup investments are going to the WRONG zones, i.e. areas workers don't want automated or that have low capability. Image
Read 17 tweets
Jun 14
The world's leading AI research center completed the most comprehensive study ever on kids and AI.

They surveyed 1,800+ children, parents, and teachers in UK.

Here's what they found:

(spoiler: children are outsmarting adults on AI) Image
The Alan Turing Institute surveyed 780 children aged 8-12 and 1,001 teachers across UK schools.

Plus, they ran workshops where kids directly tested generative AI tools like ChatGPT and DALL-E.

The results reveal how AI is already reshaping childhood:
→ 52% of private school kids use AI vs only 18% of state school kids.

57% of private school teachers are aware of their students' use of generative AI for schoolwork, compared to 37% of state schoolteachers.

The private school AI divide is creating a new class system. Image
Read 13 tweets
Jun 7
BREAKING: Apple just proved AI "reasoning" models like Claude, DeepSeek-R1, and o3-mini don't actually reason at all.

They just memorize patterns really well.

Here's what Apple discovered:

(hint: we're not as close to AGI as the hype suggests) Image
Instead of using the same old math tests that AI companies love to brag about, Apple created fresh puzzle games.

They tested Claude Thinking, DeepSeek-R1, and o3-mini on problems these models had never seen before.

The result ↓
All "reasoning" models hit a complexity wall where they completely collapse to 0% accuracy.

No matter how much computing power you give them, they can't solve harder problems. Image
Read 14 tweets
Jun 3
AI is killing entry-level jobs faster than LinkedIn can post them.

Anthropic's CEO predicts 50% of entry-level white-collar jobs will be gone in 5 years.

If you're under 30, here's what you should do: Image
First, context:

• Big Tech hired 25% fewer new grads in 2024
• Entry-level tech ads down 37% year-over-year
• Legal review roles down 45%

The question is why it's getting so much worse? Let me explain ↓ Image
Junior developers used to debug code for hours → GitHub Copilot does it in seconds

First-year lawyers reviewed contracts for weeks → AI does it overnight

Entry-level analysts built spreadsheets → ChatGPT builds them instantly

But the real damage goes deeper. Image
Read 14 tweets
May 19
Most ChatGPT answers are "yes-man."

After I began using the RPT prompt technique, ChatGPT cut its wrong answers by 40% and gave me much better results.

Here's what the RPT technique is and how to use it:

(my exact prompt + example) Image
To prompt better, first understand why AI is so agreeable.

AI gets trained by human feedback (Reinforcement Learning from Human Feedback)

When we give thumbs up to answers that sound good, the AI learns to say what makes us happy.

It's pleasing us, not checking facts. Image
The fix was discovered in January 2025 by Tsinghua University researchers.

They developed "Reasoning-through-Perspective-Transition" (RPT) prompting and tested it on 12 different types of questions.

It worked much better than regular methods by making the AI argue with itself. Image
Read 13 tweets
May 15
AI is eating up consulting.

I used 3 prompts on ChatGPT to mimic a $600/hour consultant, achieving in 4 hours what takes weeks for teams.

Here's how to make ChatGPT your pocket-sized McKinsey consultant:

(my exact prompt + examples) Image
There are 3 main tasks that a consultant does:

1. Research
2. Making slide decks
3. Reading reports

I'll show you how to automate all 3 in a way that makes one person do the work of an entire team.
Task 1: Research

Go to ChatGPT, use o3 + Deep Research.

Type this prompt:

"As a Big Four consultant, execute a DMAIC-driven Deep Research
workflow in ChatGPT—Define, Measure, Analyze, Improve, and
Control data on [topic]—structuring findings in MECE-compliant
categories into a concise executive summary."
Read 14 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

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

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(