Hi,👋 we have updated the app and fixed multiple bugs. We are lacking funds, request to free user not to use Adblock. Ads are non intrusive. 😊

Carousel Studio

Repurpose X Threads into LinkedIn & Instagram Carousels

Canvas & Ratio

Choose your destination platform format


Layout Template

Choose a content structure for your slides


Preset Themes


Typography & Sizing

Title Font Size36px
Body Font Size18px
Header & Footer Size12px

Brand Kit Customization

AGENCY

Configure brand assets for headers & footers

MULTI-PROFILES (AGENCY)
AGENCY
SAVE PRESETS (AGENCY)

Outro Slide CTA

Customize your closing call-to-action slide

#1
#2
#3

Background Pattern

Source Content

Build Your Carousel

Drag and drop any post card below onto a slide, or use the quick buttons to insert content/images instantly!

Drag Post #1
Paweł Huryn
@PawelHuryn

After an interview with @karpathy, everyone is talking about what AI agents can/can't do. But an opinion without data is just a hypothesis. So, I tested 3x185 workflow executions for a market researcher agent. The results have shocked me🧵

Apply Image
Drag Post #2
Paweł Huryn
@PawelHuryn

I tested three variants: I. LLM Workflow: No agency, the entire logic carefully orchestrated. What was expected: - An LLM workflow was 2x faster (the same model) compared to an AI Agent. - An LLM workflow consumed 12x less tokens to an AI Agent. 3/185 "errors" are minor formatting results.

Apply Image
Drag Post #3
Paweł Huryn
@PawelHuryn

II. Agentic Workflow: Deterministic logic moved to the orchestration layer. More time, more tokens. 100% task success. GPT-5 (a reasoning model) consumed less tokens than GPT-4o due to better compression. None of this was surprising. But then:

Apply Image
Drag Post #4
Paweł Huryn
@PawelHuryn

III. AI Agent: Full autonomy without steps to take, just an objective I were staring at the screen. An AI agent without predefined reasoning steps succeeded 185/185 times (100%).

Apply Image
Drag Post #5
Paweł Huryn
@PawelHuryn

This is different from my previous observations for the same models: <a target="_blank" href="https://x.com/PawelHuryn/status/1953785262611284469" color="blue">x.com/PawelHuryn/sta…</a>

Drag Post #6
Paweł Huryn
@PawelHuryn

Conclusions &amp; learnings: 1. For simple use cases, we can already achieve 99%+ reliability 2. A verifier agent with a high TPR would push it even further 3. For complex or critical processes, you still need orchestration 4. Orchestration is faster, cheaper, and more reliable

Drag Post #7
Paweł Huryn
@PawelHuryn

@karpathy might be right. We might need 10 years to achieve true AI intelligence. But autonomy and reliability for most processes seem more like ~12 months away. Agree? Disagree? Let me know in the comments. P.S....

Drag Post #8
Paweł Huryn
@PawelHuryn

A. Free n8n templates I used for testing: <a target="_blank" href="https://www.productcompass.pm/p/the-ultimate-guide-to-n8n-for-pms" color="blue">productcompass.pm/p/the-ultimate…</a>

Drag Post #9
Paweł Huryn
@PawelHuryn

B. Enjoy this? - Follow me @PawelHuryn for deep researched AI &amp; PM - Share this thread with others I appreciate it! <a target="_blank" href="https://x.com/PawelHuryn/status/1981036301060096371" color="blue">x.com/PawelHuryn/sta…</a>