Building: @kindlingtech, the trust-building system for b2b startups. Co-host: @Sachin_and_Adam, making startup documentaries.
Joined October 2018
- Tweets3K
- Following2K
- Followers3.7K
- Likes9.3K
Pinned Tweet
The pace of change in AI means the question of how we use these models, and who we get them from, is becoming one of the most consequential decisions of the decade.
The open vs closed source model debate has been one of the fiercest we’ve seen this year.
People are asking questions like:
- do we want economic concentration in a few model providers?
- what tradeoffs between capability and cost are we actually comfortable with?
- what are the real safety risks?
- which path makes the most sense geopolitically?
@baseten is a $13b company at the heart of these questions. They do inference and post-training for some of the fastest growing AI companies in the world, built largely on open-source.
But they’re not purists about open source, they believe in the many model future. We were lucky to go inside their offices to ask why, and to understand how they've built a company that's scaled to $600m ARR so quickly.
Inside the episode: interviews with founder @tuhinone and researchers @mudithj and @oneill_c
Enjoy watching!
Exclusive: Inside @baseten
The $13B startup at the centre of the biggest fight in AI - Open vs Closed.
We spent a day at their SF HQ with co-founder & CEO Tuhin Srivastava, Charlie O'Neill, Mudith Jayasekara and the Base Labs team - to understand why they're building an open ecosystem.
Backed by top VCs led by @saranormous @altcap @apoorv03 @wmccarthy11 @saammotamedi @nikiscevak @lachygroom @willreed + more. Plus participation from @nvidia.
0:00 The Open vs Closed AI War
2:18 A Billion AI Requests a Day
3:43 The Best Models Are Chinese
4:06 Tuhin: Don't Rent Intelligence
5:09 Sachin & Tuhin Talk Cricket
6:02 Tuhin: The Real Bottleneck
10:15 Why Launch Base Labs?
11:35 "You Have to Have an Opinion"
12:52 Centralised Control Is the Danger
13:07 Publish Everything, Even Fails
15:08 Hugging Face & the 6-Month Gap
16:10 Why We Still Need Closed
Adam Miller retweeted
Things that last a long time:
1. Religions
2. Cities / countries
3. Companies
Here’s how to create a religion inside your company:
- have a shared base of knowledge
thanks @davidsenra for inspiring us all
Every new creative strategist at @kindlingtech gets this giftbox.
Inside it:
1. Who is Michael Ovitz - Michael Ovitz
2. A copy of the latest @colossusmag
3. Ogilvy on Advertising - David Ogilvy
4. Goodies like a Macbook & some local cultural inspiration
We're opinionated about what we read and digest. We go back to understand the lessons from the marketing world of the 50's and 60's and how Hollywood's mega agents thrived in the 80s and 90s.
Equally, every new hire gets a list of 100 top x accounts to follow and a platter of tech podcasts to listen to that will keep them busy for the next 3 years.
There's hidden secrets to find from learning about the history of the past 100 years of marketing, media and agenting. But there's signal to be found when you're scouring the x timeline every day and corroborating your analysis with a tight group every day.
Shoutout to Matt Calvetti the architect of this lovely box.
Every new creative strategist at @kindlingtech gets this giftbox.
Inside it:
1. Who is Michael Ovitz - Michael Ovitz
2. A copy of the latest @colossusmag
3. Ogilvy on Advertising - David Ogilvy
4. Goodies like a Macbook & some local cultural inspiration
We're opinionated about what we read and digest. We go back to understand the lessons from the marketing world of the 50's and 60's and how Hollywood's mega agents thrived in the 80s and 90s.
Equally, every new hire gets a list of 100 top x accounts to follow and a platter of tech podcasts to listen to that will keep them busy for the next 3 years.
There's hidden secrets to find from learning about the history of the past 100 years of marketing, media and agenting. But there's signal to be found when you're scouring the x timeline every day and corroborating your analysis with a tight group every day.
Shoutout to Matt Calvetti the architect of this lovely box.
11 rising tides in the creative + marketing economy
1. AI video / editing / motion software. Exponential curves, already huge. Higgsfield, Runway etc.
2. AI video services. Great CDs + editors using AI to make premium work at high margin. Early, slower growth, big upside. Wonder Studios, Promise AI
3. Agentic marketing (+ GEO). Wildly scalable, lots of players, lots of slop risk, little differentiation
4. Creative talent. Every company will need strategists, ADs, editors, videographers, influencers. There's fractional / recruitment plays, the scaled plays being marketplaces
5. Small, cracked agencies. Companies with cash still want craft. $5-$20m rev teams are easier than ever. Not sure there's incentive to build the next Ogilvy
6. Owned media. YouTube/social-native shows with tiny teams. Outlier scale gets you highly profitable and asymmetric opportunities like combining distribution with venture
7. Platforms. YouTube + social platforms keep winning with creative supply at ~$0. Streamers are in a knife fight on production costs.
8. Creative data. Selling the creative process to the labs.
9. Roll-ups. With dynamics changing so fast there are likely lots of roll-up opps with synergies, across multiple vectors
10. Live / IRL / community. The one thing AI can't touch and one of the most durable assets on the planet. UFC, sports, meetups etc.
11. IP-adjacent opportunities. Owned IP or acquired IP, then using tech to leverage what you can create with it. Even without AI leverage, controlling IP is always valuable
What have I missed?
AI video, AI motion design & AI editing tools are getting exponentially better, but the demand for great editors and motion designers just continues to get higher
I see a lot of x posts saying editors are cooked with all the new tools coming out…
It couldn’t be further from the truth: the demand for high-quality talent is just going up and up
I speak with dozens of high-quality agency owners every week across videography, editing and motion and they are all killing it
The people heavily using AI editing and AI motion design tools are likely people that wouldn’t have paid for / hired editors in the first place
AI editing is just so, so far off from replacing editors. For 2 years I got our internal editing team for Kindling & @Sachin_and_Adam to try every new ai editing tool on the market and there hasn’t been a single one that’s dramatically changed workflows
That being said, with the incredible advances in AI video, I do see a much greater proportion of creative work getting done by it. But it’s important to note that high-quality AI video still needs heavy human-in-the-loop
Perhaps with the demand for video going up, we just see AI video and real production markets increase in tandem
Watched Pulp Fiction (my favourite movie) at a cinema for the first time yesterday, at the boutique Golden Age Cinema in Surry Hills (former Paramount Studio HQ).
Firstly, what a damn beautiful experience it is going to the movies and focusing on an extraordinary piece of work for hours. The fact that hundreds of masters of their craft spend countless hours creating a story that you can watch in front of a big screen for $20 is forever just so underrated for me.
Secondly, what a damn good movie Pulp Fiction is. Even watching it for the fifth time, my jaw was on the floor. The soundtrack, dialogue, eccentric plot and characters, the cinematography, it all just feels perfect to me. It’s just sensationally riveting, captivating and different. My only possible critique is that it lacks the important moral messages that are usually found in the great movies.
Go to the cinema more!
imo the best critique of Pulp Fiction from James Wood of The Guardian:
"Tarantino represents the final triumph of postmodernism, which is to empty the artwork of all content, thus avoiding its capacity to do anything except helplessly represent our agonies ... Only in this age could a writer as talented as Tarantino produce artworks so vacuous, so entirely stripped of any politics, metaphysics, or moral interest."
TIL the Cheeky Pint set is actually a room inside @stripe HQ and not an Irish pub
Cheeky Pint is back! Mark Zuckerberg @finkd and Alexandr Wang @alexandr_wang came to the pub to discuss their new frontier model, Muse Wonder.
00:00:24 Muse’s breakout use cases
00:07:28 The model is the product
00:14:35 Muse Wonder
00:24:56 How frontier models are built
00:48:26 Betting $100B
01:07:21 Agentic commerce
01:19:34 OpenRouter
01:20:06 Beyond the smartphone
01:27:00 Funding
01:34:06 Why AI won’t be winner-take-all
01:40:46 Meta Superintelligence Labs (MSL)
02:08:11 Working for Mark Zuckerberg
excited for more @Sachin_and_Adam Show episodes in the AI x bio world!
A wild week for AI x biology. We're always excited by what technology can do for the world.
> @warpsciences raised a $215M Series A
> @coherenceneuro raised a $45M Series A
> @OpenAI bought Orchestra Bio to accelerate drug dev
> @RadicalNumerics made antibacterial viruses with AI
> @AnthropicAI's Claude designed scorpion anti-venom
> @nvidia open-sourced RFD4-Proteina
> @huggingface's CarbonX can chat with DNA
> @Xaira_Thera shared X-Scientist, a virtual drug hunter
Anything missing?
Just met a 21 yr old who is building a design agency which he uses for cashflow, to fund a portfolio of 6+ different business bets
The different bets are across:
- software for agencies
- sauna company
- cleaning company
- and more
The thesis for the bets are pretty simple:
- use cashflow from the design studio to fund the bets
- focus mainly on physical world businesses
- use design capability from the agency for the bets
- use AI to primarily run each bet
- when needed, place a full-time hire into a bet (capped team of 1)
- double down if one is working particularly well
I was really impressed by this structure and his hustle to build a portfolio like this
Went away feeling very optimistic for young AI-native, distribution-pilled, high-hustle founders who want to bootstrap small businesses
Adam Miller retweeted
A wild week for AI x biology. We're always excited by what technology can do for the world.
> @warpsciences raised a $215M Series A
> @coherenceneuro raised a $45M Series A
> @OpenAI bought Orchestra Bio to accelerate drug dev
> @RadicalNumerics made antibacterial viruses with AI
> @AnthropicAI's Claude designed scorpion anti-venom
> @nvidia open-sourced RFD4-Proteina
> @huggingface's CarbonX can chat with DNA
> @Xaira_Thera shared X-Scientist, a virtual drug hunter
Anything missing?
💥 Introducing Warp Sciences, a new company applying AI and robotics to help the world develop new medicines faster.
We’ve raised a $215M Series A led by @lightspeedvp to bring frontier AI out of the browser and into the lab, and build scientific capacity here in America.
I co-founded Warp with my long-time friend @dpatil, FAIR co-founder @manohar_paluri, and drug hunter Sol Reisberg. The early team is full of incredible people like @bkspears9, who helped achieve Q>1 fusion at LLNL.
so bullish on the shipping speed of the @instinct team
Adam Miller retweeted
Awesome to see @coherenceneuro raise a $45M Series A from @foundersfund and the @OpenAI Startup Fund. When we filmed our doco with @WoodingtonBen and @Elise__Jenkins earlier this year, we were blown away by the mission, progress and team.
”Cancer is just another engineering problem”
We’ve raised a $45m Series A, co-led by
@foundersfund and The @OpenAI Startup Fund, to build cancer therapy that learns. Technology is rapidly changing medicine, but data loops and rapid adjustments are still largely missing. When you start cancer treatment, knowing whether what you’re doing is working is too slow. Clearly there is a better way. Build systems to watch cancer in real time, then feed that data into a system that itself can treat the disease. This seeds a future where, unlike drugs, we develop medicine that continuously learns to improve itself. We've been building neurotechnology and BCI platforms to do just that.
@coherenceneuro
One of the most inspiring startup missions
Awesome to see @coherenceneuro raise a $45M Series A from @foundersfund and the @OpenAI Startup Fund. When we filmed our doco with @WoodingtonBen and @Elise__Jenkins earlier this year, we were blown away by the mission, progress and team.
”Cancer is just another engineering problem”
Adam Miller retweeted
We’ve raised a $45m Series A, co-led by
@foundersfund and The @OpenAI Startup Fund, to build cancer therapy that learns. Technology is rapidly changing medicine, but data loops and rapid adjustments are still largely missing. When you start cancer treatment, knowing whether what you’re doing is working is too slow. Clearly there is a better way. Build systems to watch cancer in real time, then feed that data into a system that itself can treat the disease. This seeds a future where, unlike drugs, we develop medicine that continuously learns to improve itself. We've been building neurotechnology and BCI platforms to do just that.
@coherenceneuro
Adam Miller retweeted
OpenAI is going deeper into pharma. Orchestra Bio builds AI agents that handle the complex backend processes you need to bring a drug to market. This will only get more important as discovery accelerates.
We spoke to @joshim5, co-founder of @chaidiscovery, earlier this year, and he was seeing discovery go from years to just 24 hours of compute.
We’ve acquired Orchestra Bio, a platform that helps biopharma teams coordinate the complex process of drug development.
Together, we’ll build on Rosalind, our AI models and tools for life sciences research, to streamline those operations and help new treatments reach patients sooner.
Welcome @ashokaraj_ and team!
openai.com/index/openai-to-a…
things in tech I consume every single one of:
- @InvestLikeBest interviews
- @davidsenra interviews
- @colossusmag magazines
- @etnshow daily newsletter
- @corememory episodes
- @jasonjoyride episodes