as of 10-29-25 đ
san francisco, ca
julia@julia-rosenberg.com
@_juliarosenberg
currently: experimentingÂ
previously:Â
+ ventures lead, uniswap labs
+ co-founder & ceo, metropolisÂ
+ m&a, acreage holdings
+ student, NYU
WRITING
10-29-25 birthday bot
10-24-25 rinse & repeat
08-07-25 circle of competenceÂ
09-30-24 data trespassing
08-15-24 building in vegas
For someone whoâs spent six years working in tech, Iâm ashamed to admit I rarely touched code. I took intro CS classes in college, but my early career gravitated toward strategy, storytelling, and peopleâconnecting dots across ideas, markets, and teams rather than writing the code myself.
As a founder, I always wanted to build a technical foundation, but the opportunity cost felt too high to end up as a mediocre junior dev. In 2020, AI and vibecoding were non-existent, and just setting up a local dev environment was enough to send me into a tailspin. Still, I developed a strong theoretical understanding of technical systemsâhow infrastructure, APIs, data layers, etc. worked togetherâwithout ever building them myself.
Then, this week, I (vibe)coded!
With the patient guidance of ChatGPT, I set up my dev environmentâVS Code, Git, Python, and a few terminal basics. I explored a few paths for the project, playing with various APIs, but ultimately landed on a delightfully lazy solution: a small bot that checks a tattoo artistâs schedule and texts me when new appointments open up.Â
I used Python with the Requests library to call the artistâs Acuity Scheduling API, Twilio for SMS notifications, and deployed it via a Render cron job that runs every five minutes. The script pulls availability data, parses open slots, and automatically alerts me when something changes. Itâs a simple, passive, fully functional systemâbuilt in under four hours.
While we still havenât secured my sisterâs birthday tattoo appointment, I did give the gift of an intent in a sloppy GitHub repo. <3Â
Growing up, I spent a lot of time running around the restaurant. I would sneak into the kitchen to steal slices of fresh warm bread and softened salty butter, beg my favorite server Kevin tor an extra Shirley Temple, and print 10 different versions of âforksâ from the label maker in the small back office. Despite its fun, I also worked there one summer as a hostess, cheerily greeting all of my mom's customers and begrudgingly operating under my momâs rule and order.Â
There's a few things she drilled into me from restaurant world:Â
- The customer is always right. It's your responsibility to manage reactions and curate a positive experience. One negative experience has rippling damaging effects far beyond comping a $20 salad.Â
- Your space is a representation of your qualityâas a team, product, business. There is no detail too small and no effort lost when representing your brand and experience.Â
- No one is above the dirty work. During any shift, if you ever had a single spare moment, there was one clear direction: clean the bathroom. There is always work to be done.Â
I am now plagued by these same policies in my adult life as an OCD, people-pleasing, obsessive operator. I also now have a weird fixation on restaurant bathrooms. I still find myself tidying them out of habit: wiping water off the sink, replacing toiler paper, picking missed paper towels off the ground.Â
Yes, it is the place people go to take a shit, scroll on their phones, pick food out of their teeth. It is also the most opportune moment to capture a customerâs attentionâwhere people take a breath, sit and observe, utilize a common space in solitude. Design choices, sensory opinions (light, sound, scent), hygiene practices, etc all roll into a specific customer experience. Although itâs not your core product offering, itâs a frequently overlooked moment to deliver your customer a crisp touchpoint of your product.Â
So here are some of my favorite spots, defined by their bathroom experience:
Jules, Duboce, SF
Eel Bar, LES, NYC
Pan Pan Vino Vino, Greenpoint, NYC
Rhythm Zero, Greenpoint, NYC
Baby Blueâs Luncheonette, Williamsburg, NYC
Ore Bar, Williamsburg, NYC
09-30-24
What weâre seeing now is rampant âdata trespassingââa term that seems increasingly apt. The public web is a free-for-all, allowing AI models to siphon off public data (in ethically and legally murky ways). Cases like hiQ Labs v. LinkedIn carved out a grey area where scrapers donât technically violate laws like the CFAA. But how long will this loophole last? Todayâs AI models depend on free, publicly available data, but this reliance is unlikely to endure as regulatory and economic pressures mount.Â
As AI becomes more specializedâmore domain-specific, more "intelligent"âthe demand for both general knowledge and curated, private datasets will surge. Weâre already familiar with information companies turning their data pipelines into a lucrative commodity (from Pinterestâs 2015 Promoted Pins to Xâs 2023 API access shift). Traditional data sellers are wrestling with pricing strategies and usage boundaries as AI companies strike deals with media firms to secure vast swaths of private data for training. This is just the start of increasing data privatization and monetization.
So, what about our personal data? The debate over user privacy is, at its core, a conversation about data monetization. In isolation, our data may seem worthless, but in a diverse collective, itâs invaluable. Companies like Vana are enabling data portability and consumer data monetization, allowing users to self-upload their data or contribute to data collectives. Will we ever reap the benefits of our digital trails? Maybe. Thereâs plenty of room to further privatize (and thus monetize) internet data, but itâs complicatedâdata scraping disruptors, complex pricing models, digital ownership precedents, etc.
And what about the free, open internet? The web, once a digital commons, is colliding with the growing monetization of data. This isnât about ad-targeting anymore. Weâre entering a realm of AI models that anticipate and shape human behavior on a level far beyond todayâs algorithms. The new competitive edge is the quality of data fed into the machineâbetter data in = better data out. As AIâs appetite for data grows, so too will the conflict over ownership, profit, and access. The challenge isnât just technicalâit's legal, monetary, and ethical. The future of AI will shape and be shaped by what we feed it, and the question is no longer just how much data, but more so what data, and who provides it.
08-15-24
For nearly four years, I immersed myself in the world of crypto, co-founding and leading a venture-backed startup right out of college. It was a whirlwind of innovation, ambition, and, yes, speculation.Â
Both Vegas and crypto are elaborate human pinball machines. You launch yourself at exciting targets only to be violently bounced around a closed simulation. The infrastructure perfectly pours thousands in foot traffic through a hazardous pinball course. People come from all over to play the games: seeking status, making it big, having fun, simply fucking around. The ballâyour attention, assets, intrigueâinevitably falls through the pinball course holes. But the thrill of being close enough ensures you give it another go.Â
Both environments are designed to captivate, to excite, and sometimes, to disorient. They're realms where fortunes can be made or lost in the blink of an eye, where innovation and risk-taking are celebrated, and where the lines between game and reality often blur.
The parallels:
- Casinos are blockchains (Ethereum, Solana) competing for foot traffic, each offering their own unique gaming experience. Just as casinos vie for players with different themes and amenities, blockchains compete for users with varying transaction speeds, fees, and smart contract capabilities.Â
- Poker chips are cryptocurrencies (BTC, ETH, USDC) used to play the games. Like casino chips, these digital assets have assigned values and can be exchanged for different denominations or cashed out.Â
- Casino cashiers are exchanges (Coinbase, Uniswap) where you convert your "real-world" money to enter the game. Traditional exchanges like Coinbase act as the main cashier's cage, while decentralized exchanges like Uniswap are more akin to chip-swapping between players at the table.
- Games (blackjack, slots) are the apps (DeFi protocols, NFT marketplaces). Each offers a different risk-reward profile and complexity level. DeFi protocols might be compared to poker or blackjack, requiring skill and strategy, while NFT marketplaces could be seen as the crypto equivalent of slot machinesâeasy to play, with the allure of a big, flashy win.Â
These ecosystems offer countless variations of essentially the same games. To play the game most effectively, your job is to know what tables are hot and offer you the best chance of inserting yourself into a lucky break.Â
Newcomers, in both worlds, often don't grasp the rules. They play because everyone else is but donât understand the vigilance required to effectively engage. Itâs an insiders game: know the dealers, know the players, know the odds. You might be the lucky one who hits it big, but more likely, you'll sit next to a jaded veteran who tosses you a chip to keep you engaged.
I had a front row seat to the inside track â which tables to watch, who the big players were. But I often lost sight of the bigger picture:
I was building in a desert with limited natural resources that thrived off its unique regulatory edge/uncertainty. [Just as Vegas originally thrived in its unique gambling legal carve-outs, crypto has uniquely operated in a grey regulatory space.] We pumped water in from miles away to create the allure of abundance, attracting flashy spectacles to make their debut. The frenzy similarly operates on a 24/7 schedule that ensures a nonstop cycle of chaos.Â
[overdone young founder trope smh] I was young and bushy-eyed, and genuinely believed the potential to evolve the space beyond the speculatory games I saw around me. But the reality was sobering. The tools we built rarely shifted where the chips ultimately fell.
While I believe Crypto has immense potential as an agnostic assurance mechanism, capable of backing assets, identities, and relationships. It could disrupt intermediaries that currently dictate the rules [insert boring altruistic jargon]. But today? It's mostly a high-tech circus.
This isn't a hit on crypto, nor is it a glorification. There is still massive opportunity to âbuild rational business models around irrational attractorsâ (line from one of my favorites: Jonathan Wu). The current crypto industry boils down to two things: building the game or building infrastructure for the gameplay. Itâs still exciting to participate in a market with such strong, alluring, product forces. But those promising it's more than that currently? They're probably the ones constructing the pinball machine.Â
02-01-24
Platforms prevent us from experiencing difficult content in a healthy way. We donât have to fight through something, be patient, think so much about what something is doing to us, or create our own opinion of it as it develops because we always have that possibility of clicking away.Â
Boredom doesnât exist, difficulty doesnât exist, scarcity doesnât exist. When youâre surrounded by infinite possibilities, you can simply move on.Â
The experience of going to the movies is not about the size of the screen but the darkness of the room. It creates a fruitful sensory deprivation tank to sit with one piece of content in isolation.Â
âTo the extent taste is a kind of attunement to yourself, then developing taste in a space where it is harder to attune to yourself, where your self is more drowned out and itâs easier to jump away from that internal experience, that is also going to degrade, it seems to me at least, the fundamental ground on which that facility is developed.â - Ezra Klein
Iâm repeatedly struck by the raw, unbridled sense of freedom and indulgence seen across childrenâs artwork.Â