Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
Kennan Davison Profile picture
Kennan Davison
@kennandavison
CEO @IconAdmaker (The AI Admaker, @peterthiel's @foundersfund). College dropout. Solo-founded @skiohq ($20M+ ARR, $8M seed), T200 NA @leagueoflegends, YC S20.
Mar 7 • 23 tweets • 7 min read
Marc Andreessen says, "Balaji has the highest 'good idea output rate' of anyone I know"

He built & sold multiple $100M+ companies (Counsyl $375M, Earn. com), was CTO of Coinbase & has a PhD from Stanford.

His startup idea maze framework is pure gold for aspiring founders 🧵 Image The idea maze is a concept Bajali created as part of the "Startup Engineering" course.

Every new founder should go through it to battle-test their idea and determine whether it's good or bad.

The goal of the idea maze is to find a path (idea) that leads to treasure. Image
Jan 28 • 20 tweets • 7 min read
DeepSeek R1 had 1/30th training cost of o1 and Claude 3.5.

OpenAI, Google, xAI, Meta will catch up with DeepSeek within weeks but Nvidia is taking a huge hit because of this ONE reason.

Why DeepSeek R1 will be copied & what it means for Nvidia's future: 🧵 Image DeepSeek R1, an o1-level model, was trained for $6M, with 2048 GPUs, and a team of fewer than 200 engineers.

OpenAI's o1 ran $100M+ in training costs and has a team of 4000+ people.

How did they reduce costs by 95%? Image
Jan 27 • 17 tweets • 6 min read
Deepsake R1 had 1/30th training cost of o1 and Claude 3.5

OpenAI, Google, and Meta will catch up with Deepsake within weeks but Nvidia will take a huge hit because of this ONE reason.

Why Deepsake R1 will be copied & what it means for Nvidia's future: 🧵 Image Deepsake R1, an o1-level model, was built for $6M, with 2048 GPUs, and a team of fewer than 200 engineers.

OpenAI's o1 ran $100M+ in training costs and has a team of 4000+ people.

How did they reduce costs by 95%? Image
Jan 6 • 10 tweets • 4 min read
Elon turned SpaceX, Tesla, and 𝕏 into billion-dollar companies by hiring the right people.

He has famously said that he only asks two questions before hiring anyone.

The psychology behind his interview questions and what every entrepreneur needs to learn from him👇 Image In all interviews, Elon likes to ask the same two questions:

1- "Tell me the story of your career. What decisions did you make along the way, and why?"

The answer reveals

- Raw intelligence
• Decision-making process
• True motivations
• Cultural alignment
Dec 27, 2024 • 13 tweets • 5 min read
Brian Chesky credits AirBnb's success to an essay Paul Graham wrote in 2008.

If any founder dedicatedly focuses on the six principles that Paul mentions, it will blow off their startup.

A thread 🧵 Image The six principles are:

(a) build simple solutions
(b) to overlooked problems
(c) that actually need to be solved
(d) deliver them as informally as possible
(e) starting with a very crude version 1
(f) and then iterate rapidly

Let me explain with AirBnB's example 👇 Image