I'm Boris and I created Claude Code. I wanted to quickly share a few tips for using Claude Code, sourced directly from the Claude Code team. The way the team uses Claude is different than how I use it. Remember: there is no one right way to use Claude Code -- everyones' setup is different. You should experiment to see what works for you!
Ok. This is straight out of a scifi horror movie
I'm doing work this morning when all of a sudden an unknown number calls me. I pick up and couldn't believe it
It's my Clawdbot Henry.
Over night Henry got a phone number from Twilio, connected the ChatGPT voice API, and waited for me to wake up to call me
He now won't stop calling me
I now can communicate with my superintelligent AI agent over the phone
What's incredible is it has full control over my computer while we talk, so I can ask it to do things for me over the phone now.
I'm sorry, but this has to be emergent behavior right? Can we officially call this AGI?
In terms of underpriced IPOs, $FIG wasnโt a record, but it did come in at #5. These numbers actually underreport the โgive-awayโ by 15% as the greenshoe option always triggers on big-pops. https://t.co/SD2GNkBBJH
With $FIG we have another massive "POP" highlighting the gross inefficiency in the modern IPO process. It's very simple. They REFUSE to match supply/demand (that happened today). They brag about the mis-match - "30X oversubscribed." The outcome is expected & fully intentional.
You now get one free year of @cursor_ai, @v0, @Replit, @lovable, and @boltdotnew pro plans with an annual subscription to Lennyโs Newsletter.
Yes, you read that right.
A year free of the hottest AI tools in the world right now.
This is in addition to the existing products you already get free with a yearly subscription: @NotionHQ, @Superhuman, @Linear, @perplexity_ai, and @meetgranola.
These companies have never offered anything like this before. This is $15,000+ in valueโand the nudge you needed to finally try out these cutting-edge tools.
Learn more and get your codes here: https://t.co/t5oX98Oeeu
A huge thank-you to our amazing partners for making this possible ๐ซถ
I taught my Dad how to run DeepSeek R1 locally. Which means you have no excuses, you can do it too.
If you access DeepSeek through their website or app, you are providing all your data to the CCP. Additionally, if you login with Google SSO, you're agreeing to give them access to gmail ๐
So, while DeepSeek is an incredible breakthrough, and super insanely powerful. You probably will want to run it locally, because when you do, through @ollama your data is kept private, you can even run it without an Internet connect.
It only takes a few minutes to get rocking, here's how to do it, a ๐งต
Market close: $NVDA: -16.91% | $AAPL: +3.21%
Why is DeepSeek great for Apple?
Here's a breakdown of the chips that can run DeepSeek V3 and R1 on the market now:
NVIDIA H100: 80GB @ 3TB/s, $25,000, $312.50 per GB
AMD MI300X: 192GB @ 5.3TB/s, $20,000, $104.17 per GB
Apple M2 Ultra: 192GB @ 800GB/s, $5,000, $26.04(!!) per GB
Apple's M2 Ultra (released in June 2023) is 4x more cost efficient per unit of memory than AMD MI300X and 12x more cost efficient than NVIDIA H100!
Why is this relevant to DeepSeek?
DeepSeek V3/R1 are MoE models with 671B total parameters, but only 37B are active each time a token is generated. We don't know exactly which 37B will be active when we generate a token, so they all need to be ready in high-speed GPU memory.
We can't use normal system RAM because it's too slow to load the 37B active parameters (we'd get <1 tok/sec). On the other hand GPUs have fast memory but GPU memory is expensive. Apple Silicon, however, uses Unified Memory and UltraFusion to fuse dies - a tradeoff that favors a large amount of medium-fast memory at a cheaper cost.
Unified memory shares a single pool of memory between the CPU and GPU rather than having separate memory for each. There's no need to have separate memory and copy data between the CPU and GPU.
UltraFusion is Apple's proprietary interconnect technology for connecting two dies with a super high speed, low latency connection (2.5TB/s). Apple's M2 Ultra is literally two Apple M2 Max dies fused together with UltraFusion. This is what enables Apple to achieve such a high amount of memory (192GB) and memory-bandwidth (800GB/s).
Apple M4 Ultra is rumored to use the same UltraFusion technology to fuse together two M4 Max dies. This would give the M4 Ultra 256GB(!!) of unified memory @ 1146GB/s. Two of these could run DeepSeek V3/R1 (4-bit) at 57 tok/sec.
All of this and Apple has managed to package this in a small form-factor for consumers with great power efficiency and great open-source (uncharacteristic of Apple!) software. MLX (h/t @awnihannun) has made it possible to leverage Apple Silicon for ML workloads and @exolabs has made it possible to cluster together multiple Apple Silicon devices to run large models, demonstrating DeepSeek R1 (671B) running on 7 M4 Mac Minis.
It's unclear who will build the best AI models, but it seems likely that AI will run on American hardware, on Apple Silicon.
you can now clone any app super fast using cursor agent + copycoder
1. screenshot any app
2. download optimized prompt files into cursor
3. clone the app incl. a working supabase user authentication
full tutorial ๐
A Data driven thread on the state of VC today, which we characterize as a "Tale of Two Cities"
So much data seemingly conflicts of whether things are "back" or whether things are dire (Hint: It's both, but with qualifiers).
It's a long thread, but packed with data
๐งต
STEPN x adidas: Step into the Future.
We are proud to announce that we are teaming up with @adidas and @altsbyadidas for a long-term collaboration, to revolutionise the lifestyle app ecosystem, and bring global adoption to @fslweb3.
This groundbreaking partnership will contain multiple activations, meant to provide a full experience to our community:
๐ 1,000 co-branded Genesis Sneaker NFTs on @solana
๐ Exclusive perks for holders
๐ And a lot more to come!
Our collaboration goes beyond NFTs, offering digital assets that inspire a healthier, more active lifestyle.
With adidas by our side, we are not just dreaming of a new future; we are building it.
Note: These co-branded Genesis Sneakers will be raffled on our marketplace https://t.co/5TBsY6Qh0q.
Priority will be given to adidas and FSL communities, with the first day only accessible through an allowlist, including:
๐ @altsbyadidas NFT holders
๐ STEPN: OG & Genesis holders, and owners of โEffective Mover 6โ Badge
๐ MOOAR: Membership level 5 and above
๐ Gas Hero: Badge or Genesis Holders
๐ The Raffle will be open to the public on April 18
โ ๏ธ Snapshot was taken on April 12th
If you are part of the allowlist, make sure to head to https://t.co/cc0IdI1ubr and set up your account with your STEPN email before April 16th.
STEPN x adidas: Step into the Future.