Juan @KeyboardSloth

🇨🇦 Joined November 2018

Tweets

2K
Followers

187
Following

967
Likes

2K

AshutoshShrivastava @ai_for_success

2 days ago

Story of every CEO right now 😂

330 3K 24K 2.0M 3K

Download Image

Matt Van Swol @matt_vanswol

2 days ago

I have been struggling to comprehend why leftists are defending radical Muslims. They are essentially antithetical to everything the woke stand for and it made no sense at all. Then... it dawned on me 🧵

6K 15K 82K 9.0M 14K

Download Image

Tom Dörr @tom_doerr

4 days ago

github.com/karpathy/llama…

0 1 3 2K 4

Rohit Kumar Tiwari @_rohit_tiwari_

5 days ago

The most comprehensive, LLM architecture analysis I've read. Covers every flagship model: 1. DeepSeek V3/R1 2. OLMo 2 3. Gemma 3 4. Mistral Small 3.1 5. Llama 4 6. Qwen3 7. SmolLM3 8. Kimi 2 9. GPT-OSS Great article by @rasbt🙌 Link in the comments 👇 ♻️ Repost if you…

9 101 497 46K 580

Download Image

Brian Roemmele @BrianRoemmele

7 days ago

Turbocharge Nvidia AI GPUs: Simple Tricks for Blazing-Fast Performance In my garage I have to make what little non-VC investor research and development, work, I have pennies and get higher speed than companies with billions. This constraint causes me to find ways to squeeze more…

Brian Roemmele @BrianRoemmele

2 weeks ago

10 18 152 43K 39

Download Image

14 37 220 37K 53

Download Image

Reads with Ravi @readswithravi

2 weeks ago

THIS:

32 3K 20K 811K 6K

Download Image

Ian Nuttall @iannuttall

2 weeks ago

Claude Code 1.0.86 has a new /context command to visualize your context window and token usage It doesn't seem to be accurate yet compared to the auto-compact notice, ccusage, or my status line, but it also shows MCP tool use and CLAUDE.⁠md memory files so could be v useful!

33 116 1K 97K 642

Download Image

Full-Stack Founder @tryfoundergg

2 weeks ago

🚨 Claude Code MAX 20x - has 1 million context window:

46 87 1K 324K 471

Download Image

ℏεsam @Hesamation

3 weeks ago

this 3b1b visualization of how LLMs work layer by layer is the true brain rot we never knew we needed.

45 611 6K 287K 4K

Download Video

Brian Roemmele @BrianRoemmele

3 weeks ago

HUGE AI/Cancer Breakthrough AI automatically designs optimal drug candidates for cancer-targeting mutations. KAIST researchers developed a game-changing AI model, BInD, that designs perfect drug candidates using only the target protein’s structure—no prior molecular data needed.…

36 123 578 75K 192

Download Image

Seb Vidal @SebJVidal

3 weeks ago

Here it is! What's New in UIKit in iOS 26. This was a pretty mammoth undertaking, and it shows – a whopping 64 minute read! I'll be keeping this article up to date over the beta cycle. Possibly littered with errors. Shout at me if I've made any stupid mistakes. Link below 👇

31 40 445 55K 280

Download Image

Juan @KeyboardSloth

4 weeks ago

0 0 0 31 0

Download Image

eric provencher @pvncher

4 weeks ago

It’s interesting seeing gpt 5 in use in different places- it’s currently many different experiences - In ChatGPT, 5 feels like a step down. The router doesn’t work, and you talk to dumber models more often - gpt 5 high over api is a really great model though, but almost no one…

18 4 108 13K 38

ℏεsam @Hesamation

4 weeks ago

this is the most comprehensive resource i found so far which gathers all the key things to know about Context Engineering. Absolutely beautiful work.

9 93 1K 97K 2K

Download Image

Kieran Klaassen @kieranklaassen

4 weeks ago

Claude Caude can run GPT5. GPT-5 is good at fixing nasty bugs and doing research. In a different way than Claude and they can work together. Just create a Claude code agent called gpt5:

72 65 981 117K 1K

Download Image

Ian Nuttall @iannuttall

4 weeks ago

I just compared Claude Code vs Codex vs Cursor CLI The task was to build a Next.js app with Tailwind 4 and shadcn components to collect customer feedback and showcase it with a widget. I gave all three the same prompt and let them go for 30 minutes to see what they came up…