Rob Imbeault✓
@RobImbeault
Backboard CEO | Unicorn Founder | Author
567 Following2K Followers7.3K Posts
- I've gotten addicted to this app.Sanzio @BackboardDevsThis is why GPUs are $6000. Some of the stuff building with H3 right now is wild. 1 movie scene, 3 drivers, every face swapped. Same camera, same timing, same audio.00386
- Throughput matters to the operator. Waiting matters to the user. In our GLM 5.3 burst test on 8xB200, median time to first visible answer was: @BackboardIO: 66.2 seconds vLLM: 283.6 seconds Still room to improve. A meaningful difference already.
01172 - I knew it!Shane Parrish @shaneparrishADHD is a superpower. "The problem isn't that I can't focus; it's that I can't focus on stuff I don't care about." "On stuff I care about, I can lock in and you’re not going to find me for months." So if you have ADHD ... "The first thing you have to do is reframe the task to find the autonomy."004269
ADHD is a superpower. "The problem isn't that I can't focus; it's that I can't focus on stuff I don't care about." "On stuff I care about, I can lock in and you’re not going to find me for months." So if you have ADHD ... "The first thing you have to do is reframe the task to find the autonomy."551881.9K155K- Rob ImbeaultRT @awilkinson: AgreeDick Lucas 🇺🇸 @dickclucasIf you pay $35k for a vehicle and it doesn’t drive itself you have failed the iq test and put your family at risk.0401
If you pay $35k for a vehicle and it doesn’t drive itself you have failed the iq test and put your family at risk.
1121131.6K92.5K- What happens when the workload keeps coming? In our GLM 5.3 sustained test at 32 concurrent users: @BackboardIO: 370 output tokens/sec vLLM: 74.9 4.94X aggregate output throughput. Full model. No Quant!
235243 - "sf is just a waterloo wrapper." love it!Austin Walker 🛴 @austinxwalkersf is just a waterloo wrapper. let's keep it going! i'm hiring an AI-native software engineer to our founding team. looking for a canadian new grad who's hungry to build, move fast, and own entire parts of the product from day one. will cover your relocation to SF + visa. we’re building Atlas, a personal health agent for the 500M people living with chronic conditions that have no approved treatments. you’ll work directly with me to help build the future of patient-led healthcare. DM me your github + something you’ve built recently!202303
sf is just a waterloo wrapper. let's keep it going! i'm hiring an AI-native software engineer to our founding team. looking for a canadian new grad who's hungry to build, move fast, and own entire parts of the product from day one. will cover your relocation to SF + visa. we’re building Atlas, a personal health agent for the 500M people living with chronic conditions that have no approved treatments. you’ll work directly with me to help build the future of patient-led healthcare. DM me your github + something you’ve built recently!
23210813.3K- You should probably go to this.Lucy Hargreaves @lucyhargreaves4Join me in Ottawa on Wednesday with special guest @notnotaspy106259
- now if you'll excuse, i have to upload anthropic's s-1 to anthropic so it'll give me something to tweet about anthropic00057
- Our latest GLM 5.3 full-model BF16 test: @BackboardIO: 165.5 output tokens/sec vLLM: 63.5 output tokens/sec 2.61X aggregate output throughput. Same 96 workflows. Same request bodies. 32 concurrent users.
036372 - It’s already startednir doshi @Nir_Doshi1For all the news about neoclouds on being on the verge of collapse, @BackboardIO is like a knight in shining armour, we can save them so much that they wouldn’t even have imagined that.001121
For all the news about neoclouds on being on the verge of collapse, @BackboardIO is like a knight in shining armour, we can save them so much that they wouldn’t even have imagined that.006175- Rob ImbeaultRT @martin_casado: Why Hand Wringing about Defensibility Is Probably Pointless: - analyzing how product value grows as a company scales is really hard -- this generally involves reasoning about conversion properties of an adaptive feedback system - analyzing how product value is impacted by competitors' scale is really hard -- also adaptive feedback system - analyzing competitive equilibrium is really hard -- autonomous differential equation - stochastic nature of serendipity and setbacks are really hard to bound - all this shit is dynamic based on macro factors, team, cyber threat environment etc. - market growth trumps all this stuff anyway, as fast growing markets create new white space which totally changes the value functions I've now read a lot of the key papers on this stuff, and most of it is either post hoc, or makes mature market assumptions which I just don't think are all that relevant with early stage tech startups. The tl;dr is invest in fast growing markets.012012.6K
- This is a euphoria day. I’ve decided to enjoy it before the terror sets back in.00261
- Wonderfully articulated. 🙌Brad Gerstner @altcapBeautiful & true. 🎯🇺🇸🤍00057
- When I say I’m going to manifest something what I really mean is that I’m going to engineer it.00282
- Show more posts
