chris muscarella
@cm
Followers
8K
Following
34K
Media
152
Statuses
9K
Instigator. Trying to learn. Africa / USA. Partner @timoncap Iron Man @thefieldcompany
Brooklyn, NY
Joined December 2006
Coda: there’s good arguments about how ETFs / index funds have made markets weird because of lack of active investors doing ‘opposition research’ and curious about implications of turning the dial to 11.
0
0
1
Pretty sure this + Robinhood’s announcements are beginning of end for ETFs…. Of course in future you just own tokenized securities and smart contracts auto balance positions and loss harvesting….
This produces virtually the identical performance as buying a market ETF. However, since I own each name individually, I will have more individual losses at any time. And by "harvesting" these losses, I get close to 30%-40% of my initial investment as a tax deduction!
1
0
1
The important caveat: it's not all about price. Fundamentally stablecoins expand the market / push to "Internet Native Finance" in terms of who has access to FX and how frictive it is. ie, don't think of the current market as the actual market cc: @NikMilanovic.
0
0
4
Don't fully agree with everything Jack from Airwallex is saying. but as someone who's traded nine figures of FX in Nigeria starting in ~2019 can tell you stablecoin pricing has gone from shooting fish in barrel to worse pricing than regulated remittance inflows. .
Investors keep asking me about stablecoin, and how that can reduce FX fees; if you send money from USD to EUR, and the receiving end still requires to receive EUR in their bank, I can’t see any ways stablecoin can reduce fees - off ramping from stablecoin to recipient currency.
3
0
3
And that there's a number of really interesting ways to do that curious who's seen interesting approaches?.
0
0
1
. and Tutor was also the company that shocked me a few years ago by using transformers to shrink training time by ~80X around the launch of the GPTs. Think the approach of sidestepping giant "systems integrations" in whatever way is likely correct.
1
0
2
Very curious to watch this development path and whether there's elements that rhyme to the "start out looking like a toy". Was an early investor in Tutor Intelligence which started with robotic arms only. .
Will the controls that allow humanoids to take on general tasks also be such that integration in factory robotics will shorten from the current year or more? Those delays might be structural not technical.
1
0
3
As someone with a distinctive writing style and meter that loves the em dash—slightly miffed at AI slop. Not the same kind of flattery as a human impostor.
0
0
2
An excellent thread on the whiplash of tariffs in consumer goods. And another way in which I'm grateful for the second time (Covid was first) that the Field Co has an almost completely domestic value chain.
It's all a bit chaotic and unthoughtful. We cannot execute on the behavior tariffs were created to encourage (re-shoring or friendshoring) fast enough to make a dent in this year's P&L. Instead, we're resorting to price. /10.
0
0
1
Next phase: stablecoin issuance not being a great business ((give up yields to distribution partners / users) * (current U.S. admin wanting to push down treasury yields)). Phase after: programmable currency / lots of AI agents.
@SamoraKariuki @__JasonMarshall @AyowoleOA Hot take: stablecoin rails as a means of payment rails are overestimated by VCs: stablecoin tied into rewards programs / proprietary distribution networks are wildly underrated.
0
0
0
A good proxy for how much someone understands what's going on: what's their take on how stablecoin rails tie into consumer remittances?.
@SamoraKariuki @__JasonMarshall @AyowoleOA Hot take: stablecoin rails as a means of payment rails are overestimated by VCs: stablecoin tied into rewards programs / proprietary distribution networks are wildly underrated.
0
0
0
And also just very interesting to see what that means around legacy brands / M&A where those companies do not have that kind of DNA. .
0
0
0
Haven't tried to go too crazy with modeling. but think you're going to see companies that can be in the 30-45% EBIT range with $50-300M in topline much more frequently.
1
0
0
Starting to really scale revenue in a consumer products company that is digital native, with a super low headcount, where folks are AI native and curious— feels like magic.
2
0
4
Do I know anyone with a mid-size team that's really good at permissioning and sharing across Google Apps / calendar / docs / etc? These tools are a nightmare.
1
0
1
Much more interested in the next gen of platforms like ServiceTitan where the value is built more as a co-op and better AI tooling for small business owners— and it’s not for PE shops.
🚨ServiceTitan S-1 Out!🚨. $685M Revenue (up 24% YoY).$62B Gross Transaction Volume (up 23% YoY).>95% Gross Dollar Retention.>110% Net Dollar Retention.77% Non-GAAP Platform Gross Margins.72% Platform Gross Margins.$(183M) Net Loss (31% YoY Improvement). Analysis to come.
0
0
1
Any recs for world class hard drive recovery services (NYC great, but can ship)? Particularly that can handle a Macbook with a disk that's integrated into a fried M2 motherboard that functions as a unit?.
0
1
3
Interesting parallelisms around biotech companies effectively using ML to do massively time condensed cross breeding / genetic selection instead of direct gene modification (gmo / crispr) and some of Jensen Huang comments that in future all images generated not rendered.
0
0
0
“Wild computation” and implications around mutations.
Stephen Wolfram says there may be no way to readily identify and explain the mechanisms behind machine learning because, like neuroscience and biology, it may depend on "wild computation" and be the outcome of computationally-irreducible evolutionary adaptation
1
1
1
Not a shocker: machine learning in many ways showing signs similar to adaptive biological evolution (video of Wolfram hitting gist in reply).
What's really going on in machine learning? Just finished a deep dive using (new) minimal models. Seems like ML is basically about fitting together lumps of computational irreducibility . with important potential implications for science of ML, and future tech.
1
0
1