Margatix

494 posts

Margatix

Margatix

@margatix

angel investor, founder & generally techno-optimist

Katılım Mart 2022
401 Takip Edilen82 Takipçiler
Tykoo
Tykoo@0xTykoo·
If you’re a founder or tech investor planning to visit China, feel free to DM me. I’m based mostly in Shenzhen and spend most of my time around AI investing, agents, hardware, payments, trading, social products, and AI-native consumer toys. China is hard to understand from the outside. A lot of the most interesting things are not obvious from Twitter. If you’re building or investing in: - AI agents - agent payments - A2A platforms - AI trading - social products - sports / consumer AI devices I’d be happy to connect, show you around, or introduce you to relevant people. When you DM, just tell me: when you’re visiting, whether you’re fundraising, and who you want to meet.
Tykoo@0xTykoo

Network School China Tour, Shenzhen I’ve brought a bunch of friends (founders, developers, engineers and designers) from @ns to China, first stop Shenzhen⬇️

English
48
1
40
7K
Margatix
Margatix@margatix·
4 days in Shenzhen and can confirm, don't sleep on it in many ways it genuinely feels like time travel to the future > supplier response times are in minutes! even on weekends, because everyone actually wants to work and get business > robots deliver your food, high-end electric cars are the default, one-tap payments work across every merchant, power banks at every corner > streets are clean and safe without being overly sterile (ppl playing instruments, kids playing, etc) > everything you need to live and function is inside a 15-minute walk > the entire region is 1-2 hours from major hubs like hong kong, guangzhou by high speed rail its 'urban life' optimized for doing things (rather than preventing them) and the hardware ecosystem is just the most legible expression of that
Margatix tweet mediaMargatix tweet mediaMargatix tweet mediaMargatix tweet media
Zac Valles@zacharyvalles

72 hours after YC demo day, I moved to Shenzhen for 8 weeks 🤠 I'm headed back to SF with new hardware in hand (sharing more soon), but some takeaways documented below: > If you have even the slightest ambition to found a hardware company, visit SZ. Pre-raise, pre-team, pre-idea, pre-job departure, it doesn't matter. Just go. > Plan your visit according to a major conference that interests you. Use that conference as a supplier meeting springboard - that's your ticket to any factory under the sun. > At the factories, ask about lead times, don't ask about cost (wait on this). Your iteration rate is driven by the lead time on the longest lead time item in your assembly. It pays to identify these parts early to build project timelines. > Visit Huaqiangbei (read: this is a mini-city, not a building). Robotic subassemblies, batteries, chassis's, electronic parts. They all have buildings where vendors are tightly clustered. Plan to spend 4-6 hours walking around before you find exactly what you're interested in. > Business relationships are valuable commodities. Treat them as such. Pay attention to people, learn about them. Bring thoughtful gifts. Wait for them to sit first. With Baiju, fill the glass but with tea leave some room. Cultural customs are fun to learn, but also convey a seriousness towards the working relationship. > Suppliers fit cleanly into discrete buckets. Level of complexity and execution on past projects indicates what is in scope for them. Trivial, but important to level your build expectations. It is easy to design a part with 12 subsequent manufacturing processes, exceptionally hard to find a supplier to fill this order. If you need coffeeshop recs, food recs, or hotel recs I have a few. Move to Shenzhen! Get to building!

English
0
0
6
94
Margatix
Margatix@margatix·
there is some nuance depending on where / from who people bought their secondaries a lot of the original buyers bought the shares through LLCs (and then people bought / exchanged the ownership in those LLCs) what actually trades on the secondary market is often the membership interests in those llcs (and not the underlying stock itself) so on anthropic’s books nothing changes, the same llc remains the holder of record, just owned by different people now piercing a legitimately formed holding llc is a very different legal fight than voiding a direct transfer ​​​​​​​​​​​​​​​​and arguably much harder for anthropic to enforce
English
3
1
39
4.4K
Margatix
Margatix@margatix·
yeah and this is just one of the nuances another is that there is a really meaningful difference between the type of shares people were buying on the secondary market e.g. the Anthropic Series B shares sold in the FTX bankruptcy sale came with a ROFR waiver and about 80% of those buyers purchased these shares with LLCs so whoever owns % in those LLCs = owns the underlying and piercing a legitimately formed holding llc compared is a very different fight than voiding a direct transfer​​​​​​​​​​​​​​​​ nearly impossible for Anthropic to enforce
English
0
0
3
616
Thomas Braziel
Thomas Braziel@Bkclaims·
The Anthropic/SPV situation is way more legally complicated than “the transfers violated the charter, therefore buyers get zero.” That is not how Delaware equity jurisprudence works - especially in Chancery. Sure - Anthropic can absolutely argue that transfers required board approval and that certain structures attempted to circumvent transfer restrictions. Fine. That is a serious argument. But Delaware courts also care deeply about acquiescence, waiver, estoppel, reliance, and equitable fairness. And that is where this gets messy. These secondary/SPV structures did not appear overnight. This ecosystem existed openly for YEARS. Deals were marketed publicly. Prices were tracked publicly. Entire platforms existed around them. There are almost certainly internal emails, texts, compliance discussions, board materials, screenshots, and executive conversations acknowledging these markets existed and choosing not to enforce against them in real time. At some point Chancery starts asking uncomfortable questions: Did the company knowingly allow a secondary ecosystem to develop? Did sophisticated parties rely on that silence? Did insiders themselves participate in or benefit from these markets? Did the company selectively enforce only after valuations exploded? Did they “sleep on their rights” while billions in reliance capital formed around these structures? And even IF some transfers are ultimately voidable, that still does not mean counterparties are left with no remedy. Delaware equity courts are not blind to unjust enrichment, reliance damages, constructive trust theories, rescission claims, tortious interference issues, or other equitable relief where parties acted in good faith based on years of tolerated market practice. That is the key point people are missing: this is not just a pure four-corners corporate charter case anymore. Once a company knowingly permits an entire shadow secondary market to flourish for years, equity enters the chat. And Delaware Chancery is literally the home of equity.
English
28
38
420
71.7K
Ankur Nagpal
Ankur Nagpal@ankurnagpal·
If Anthropic deems all secondary sales of Anthropic stock to be void Does that mean the original buyer retains financial interest even after selling it away? Lawsuit territory
Ankur Nagpal tweet media
English
27
10
214
55K
Margatix
Margatix@margatix·
@JTLonsdale it's underappreciated how strong the power law is with criminals and repeat offenders 20% of ppl causing 80% of distress but unfortunately that's a reality some want suppressed
English
2
2
58
3.5K
Margatix
Margatix@margatix·
@Seanfrank exactly despite these seeming "headwinds" there has simply never been a time with such an abundance of tools and tech that amplifies humans (our capabilities and skills) if you are destined for success, its easier and more attainable now than ever before
English
0
0
1
314
Sean Frank
Sean Frank@Seanfrank·
Want to know a secret? Despite the war. Despite the tariff mess. Despite everyone scared of a recession. Despite ads getting more expensive, Amazon being evil, and more competion. It’s actually the best time ever to be building a brand. Easiest time to win and get rich.
English
22
5
230
10.7K
Margatix
Margatix@margatix·
Why do people always have negative implicit assumptions about someone valuing their time? If you've built a successful business, you don't need to lock yourself in a room You won one type of game and are moving onto another You have capital, knowledge, experience and want to connect, support and bet on others Having a steep price point is just a natural filter to attract those who a) can afford it already (as a result of their actions) b) find it expensive but value it enough to pay fot it & make it happen Both are a great way to match the right people and align incentives
anul agarwal@anulagarwal

when you have had successful businesses why would you consult people for $2,400/hr? i don't get it?

English
0
0
2
77
Margatix
Margatix@margatix·
@levelsio @nasdaily landed yesterday all in all including taxi time, opening the door, getting through immigration, getting luggage ~11 mins AC perfectly set too (not too hot, not too cold like many SE countries have a tendency to) absolutely elite. have yet to an airport outperform
English
1
0
4
95
@levelsio
@levelsio@levelsio·
@nasdaily What's not true here? The bag throwing or the sweaty temperature
English
3
0
33
3.1K
Margatix
Margatix@margatix·
@Jason weird way to try to get in on the action Jason..
English
0
0
13
1.5K
Margatix
Margatix@margatix·
"it lets us agree on shared values and enforce them collectively" this is awesomely formulated and the one point that many in the space miss collectively deciding and agreeing how a decentralized system should function in fringe situations and protect itself is not at odds w the promise of decentralization
English
0
0
4
551
Haseeb >|<
Haseeb >|<@hosseeb·
I don't know who needs to hear this, but North Korea is your enemy. They hate what you love. You should want them destroyed. Every loss of theirs is your win. Blockchain has real enemies, and among them North Korea is the final boss. To prostrate yourself before them, to defend their right to keep attacking you, is the height of self-hatred. Decentralization is not a suicide pact. It's the opposite: it lets us agree on shared values and enforce them collectively. It means we have no higher power to answer to than each other, and no one can stop our collective will. So here's a value we should all agree on: Fuck North Korea. They don't get to win. We do.
Haseeb >|<@hosseeb

FYI, this is a validity violation, not a decentralization violation. They say governance will vote on what to do about the cordoned off assets. In principle, governance could vote to let DPRK bridge out. It's possible (in fact, core to the value proposition of crypto) for a large set of people to agree to do something that a plurality want to do.

English
109
46
579
65.3K
Margatix
Margatix@margatix·
@akothari @contraben so spot on. a little sad to see how everyone is optimizing for 'scarcity/savings' by default instead of focusing on what all these things enable us to do, achieve and how much we can *grow* the pie instead
English
0
0
2
122
Akshay Kothari
Akshay Kothari@akothari·
You can use AI to cut costs. You can also use AI to raise your ambition. Too many people focus on the first. More people should be thinking about the second.
English
91
121
1K
53.4K
Margatix
Margatix@margatix·
genuinely hard to express how useful and valuable this is and will *increasingly* become in a world where everything around us becomes automated, the last remaining moat is creativity and taste and up until now LLMs had no meaningful, scalable way to access and tap into that *human element* now they do and not because the underlying "technology" is somehow novel, but because the INPUT is professional human taste at scale is the one thing you can’t synthesize and can’t scrape contra also didn’t start as an AI company, they started as a creativity-first platform which is exactly why they’re the only ones with the network, the data, credibility and all the right synergies to pull this off they have it by default because they built for creativity long before AI made it valuable
ben@contraben

We benchmarked four leading AI models: @AnthropicAI (Claude Opus 4.6), @GoogleAI (Gemini 3.1 Pro), @OpenAI (ChatGPT 5.3 Codex), and @Alibaba_Qwen (Qwen 3.5) on real landing page design tasks. Same prompts. Real designers. Multi-phase evaluation across ideation, structure, and refinement. The results weren’t what we expected.

English
0
3
13
858
Tyrel | Natty Ice Cream
Tyrel | Natty Ice Cream@tyrelsj·
Some unfortunate news. I can't reveal which retailers @EatNatty is rolling out at in April - yet. But I can share that we'll be expanding to over 200 retail doors. Two new flavors and our new incredible formulation. 6X growth incoming. Stay tuned!
Tyrel | Natty Ice Cream tweet media
English
8
0
25
1.6K
Margatix retweetledi
Sean Frank
Sean Frank@Seanfrank·
Ai job loss- my perspective as the decision maker in a $100,000,000+ company: 1- ai job loss is over hyped. we have fired zero people because of ai we have fired people for refusing to do the work on their job description. we have fired people for being bad at their jobs. we have fired people for not trying to use new tools. But we are still set to SLOWLY increase headcount this year. maybe net 2-5 people. 2- the real effect is job growth slowing. I will always need someone who "owns" a vertical. I have an inventory director. She needs to be the decision maker in all inventory planning. Agents SHOULD help her. Agents should do most of the work. But it needs to roll to a human. If she is gone, it has to go to my COO or me. Just not functional at a large org. I see the revenue per person target just increasing. 1 million was gold standard. Now its 2m. and by 2027 it will be 5m. More work being done. the same number of humans in the loop- not less. 3- The stagnation in the knowledge work market will be offset by the demand in blue collar work. this is obviously happening right now. My brother in law is a welder. there is unlimited demand. we run a production facility. that has doubled in 12 months. There could very well be a decline in "email" jobs. but that decline will be modest, and will be more than made up for in the blue collar boom. eventually, we will build houses and mines and factories in america. 4- every 20 years, the hot job has to change. It was finance in the 80s. tech in the 2000s. and now it will be plumbing and electrician. I remember the "learn to code" movement around coal miners. It came off as mean and tone deaf. but I think we will need a program to teach former product managers from SAAS companies to learn to weld. 5- Jobs are fake. In 1800, 95% of people were farmers. In 1900, 80% of people were farmers. In 2026, 3% of people are farmers. If you went back in time to 1880, and explained your job to a coal miner, he would assume you were the god king of earth. But no, we just tippy type and order private taxi for our food. The work we do just changes. we are going through a change now. on the otherside, I think there are higher wages and lower prices for everyone. service work will be seen as high status. baristas will make 100k a year. the future will be awesome, because tech will do more and more of the work. like farming today. ---- Not financial advice, but I dont believe the doomerism. Im daily cost averaging into the market. We are going to see the single biggest quality of life jump EVER. 4 day work weeks, higher saleries, people building shit again. as you know, I am an optimist. NEVERDOOM!
Sean Frank tweet media
English
49
21
340
31.6K
Margatix
Margatix@margatix·
while US is undoubtedly the best place to do business its never-ending ability to form new extractive legal verticals around emerging technology is so tiring so much value extracted & wasted on completely non-value additive, non-productive stuff while also further detracting many from launching/operating a business
English
0
0
1
50
isaac
isaac@theisaacmed·
For years I’d watch friends get trolled by ADA lawsuits. It was easy money. Brands almost always settle. There is a whole industry of law firms that specialize in this. AI ads are gonna be the new ADA lawsuit.
isaac tweet media
English
10
0
53
7.9K
Margatix
Margatix@margatix·
I believe these people find themselves to be "artsy" precisely because they do not seek convention, but rather something unique, distinct, unusual and they find the "mass appeal" unattractive and off-putting which kind of goes hand in hand with their distaste towards capitalism that caters and optimizes for the "mass appeal" because its what makes sense in most forms of commercial arrangements (find a lot of ppl wanting the same thing, produce it and distribute at scale) so I think (on average) you'll have a hard time convincing many of these people, by the nature of all these dynamics at play
English
0
0
1
111
Sam Parr
Sam Parr@thesamparr·
I love Derek. He came on MFM. I tried to convince him + people like him that capitalism is good. And that not only can you make a great living while having fun and providing value, but if you're really good at what you do, you owe it to the world to make money so you can continue giving value to the world. Something I've noticed about people who are artsy, soulful and amazing at their craft (those are all good things)...they think making money or charging for something is bad. My goal is to convince them that they're not just wrong but they're doing their craft a disservice.
derek guy@dieworkwear

my impression is that it's very easy to make money online so long as you're willing to be a boastful, ego-centric bullshitter and sell sham products to a growing fan base. you can easily be a millionaire this way. and perversely, this wealth only earns you more credibility

English
75
9
482
101.5K
Margatix
Margatix@margatix·
@yongfook latter but AI tools enable the above-average worker to produce 10x more output so it all balances out in the end and the barbell of output tied to one's intelligence widens even further
English
0
0
0
69
Jon Yongfook
Jon Yongfook@yongfook·
What is more likely? AI tools enable the average worker to produce 10x more output. AI tools enable the average worker to produce the same output as always, with 10x less effort. Lets be honest...
English
76
2
160
18.3K