Daniela Perlein

I feel like we’re all playing “boar on the floor”

@CohenSite @nithyavraman as a CD4 HPOZ resident I am horrified. What’s your office doing to fight this? What would you do as mayor?

LA's new proposed "objective design standards" for historic districts are horrific. They're going to result in it essentially being illegal for new buildings to match the historic character of these districts.

@bryan_johnson This is very exciting, but it is not world-first. People have done all of these things on Reddit for what it’s worth. I nonetheless champion your bringing awareness to this because it’s insane how little endometriosis has studied. Keep going!

🚨Kate officially diagnosed with endometriosis > done without surgery > across three modalities: imaging, blood, and AI > all non-invasive > in 42 days For context, average time to diagnosis is 6.6 years. And, we found 2 other diagnoses at the same time. Over the past 6 weeks, we’ve sprinted to confirm or deny Kate’s suspected endometriosis. Endo is notoriously challenging to diagnose. It’s one of the most gnarly diseases and affects 15% of women. Men, to get you on the same page, having endo is akin to an alien growing in your guts and balls, self replicating, and glueing everything together. Causing you constant pain and discomfort. We got to work. > got an MRI > got a transvaginal ultrasound > both results came back negative At this point, Kate’s patient journey had followed the archetype precisely. Most women don’t get diagnosed for 7-10 years. For Kate, it’s been 7 years. And, like most women, her imaging came back clear even though now we know that she has endo. This is why diagnosis has traditionally happened via surgery. There has been no other way than to open her up and look inside. We wanted to avoid surgery so we went back to the drawing board. We searched the world over. On our second go, we did: > endo-specific ultrasound > an endo blood test > AI MRI > saliva test This was successful. We were able to confirm her endo via ultrasound, blood test, and MRI. Confirmed simultaneously by three unique modalities, as far as we know, a world-first approach. The extensive measurement allowed us to find additional things. Her ultrasound showed: > endometriosis > PMOS (formerly PCOS) (needs confirmation) > adenomyosis 30-40% of women have at least one of these conditions. That’s intimidating especially when the path to diagnosis is fraught with so many challenges. Phase I was getting a diagnosis. Phase II is curing endometriosis. We’ve already started working. If you’re a female with suspected endo, here’s what you can do to accelerate your diagnosis. — 1. Endo-Specific Ultrasound You want an endo-specific ultrasound. As we saw with Kate, a standard pelvic/transvaginal ultrasound failed to identify her endo. You want the ultrasound to be performed by a physician or sonographer specifically trained. They follow a special protocol to hunt for endometriosis by mapping the ovaries and uterus, and testing whether organs can slide freely, or are tethered by endometriosis lesions. It’s best timed just after ovulation, when a small amount of peritoneal fluid aids visualization. It can detect superficial endometriosis, lesions, and adenomyosis that general imaging misses. We went to Dr. Kacey Hamilton at Cedar Sinai. 2. AI MRI / MatricesAI Radiologists miss lesions in up to 60% of cases. We worked with @MatricesAI which leverages AI and a unique dataset to detect endometriosis lesions on pelvic MRI. This model is still new, its first pilot study with 200 participants began in April this year. Here is how you can work with them: They’re opening their diagnosis program at the Geneviève Institute to give early access to their AI model in a clinical trial. They will take you through a state-of-the-art clinical intake questionnaire. Help you advocate for your symptoms with your gynecologist, based on your intake or connect you to a new center where their AI is being piloted and the clinical trial conducted. 3. Blood test / HerResolve Kate had two small tubes of blood drawn for a test built by @Heranovalifesci. The test measures seven biological markers (three microRNAs, three proteins, and one hormone) and uses an AI model to help detect endo. It was highly accurate in its validation study at confirming endo and caught most cases that ultrasound and MRI had missed. Their technology has been validated in a peer reviewed study (298 women, 11 sites across US/Europe/Hong Kong, published in the Journal of Minimally Invasive Gynecology): specificity 97.5%, sensitivity 80%, with strong diagnostic performance (94.4%), demonstrating it was highly effective at distinguishing women with endometriosis from those without the disease. A positive result is a strong signal, since only 2.5% of women without endometriosis test positive, though final confirmation is still clinical. A negative is less conclusive, because the test misses about 20% of true cases. The test identified 61.5% of histologically confirmed cases that transvaginal ultrasound and/or MRI missed. # We did one more saliva based test and will report back on that when results are returned.

Why can’t we have nice things?

@CohenSite @cd4losangeles as a CD4 HPOZ resident I am horrified. What’s your office doing to fight this?

A year ago, @FounderLaberge was putting sensors on tractors and excavators to catch failures before they happened. Today @TryOmenAI announced $41.5M — applying that same real-time fluid analysis to the liquid cooling inside AI data centers, where it's becoming mission-critical as compute demand compounds. Proud to back this team. They're hiring in SF & Orange County.

Very important lesson for Blue Cities here: Every change Lurie has made has been opposed by screaming, bullying activists -- ignoring them has been the path to policy success and extremely high popularity. @skaushik100 sfchronicle.com/election/artic…

“In the Los Angeles area, for example, the average exit tax for homeowners 65 and older is $185,000, while the annual cost of insurance, maintenance and property taxes on a zombie home is only about $10,000, the data showed..” nytimes.com/2026/04/30/rea…

@JoeFernandez Is there anything that can be done locally despite prop 13? There’s a house on my block that has been empty for 16 years since the owners died; their kids have let it rot — needless to say, it’s an eyesore. Occasionally coyotes nest in there.

One thing I see in my neighborhood in LA with Prop 13 is that there are at least 8 houses in a 2 block radius that have been empty for 5+ years. People who owned them have passed and it's so cheap for their kids just to leave it sitting. Bad for the neighborhood and city

@mistressdivy @default_friend nerve.com

What’s a "lost" website from the early 2000s that you still think about today?

@costasipsis @jwalkermobile Can you do this for LA?

@jwalkermobile Lol I spun up a site for this exact reason, tracks a bunch of hard to reserve spots and provides the timing of each tabledrop.nyc

I get asked fairly often on how to secure reservations at NYC's hottest & hardest to get restaurants, nothing is for certain, I do believe knowing their reservation policies, and WHEN to try, will help. Here's a list of the hardest reservations to secure, and when to try and make your reservation.

@nunzi46 @Harryoc493 <3

Congrats to the one and only @Harryoc493 on the Thiel Fellowship! Even more impressive with it being such a small class this year Couldn't have happened to a kinder person with a more authentic vision for the future

How lucky we are to experience boring, ordinary, uneventful days. Somewhere in the world, that kind of safety is unimaginable.

On our way to track @NASAArtemis using passive SONAR. Stay Tuned

@peachofthewest Knock on that door! The world needs to know 😉

@danielaperlein Oh I’ve never met them lmao I just feel the bass radiating through my floorboards every night and it’s insane

the person who lives below me in my income-restricted koreatown apartment building is a musical savant on the level of 2012 Grimes

@nunzi46 🫶

To appreciate where we’re going, you’ve got to understand where we’ve been!

living in a walkable neighborhood in LA is actually the best

Solid Curtain '26 was a counter-UUV exercise that brought together multiple mission stakeholders and demonstrating teams to detect asymmetric underwater threats in semi-contested port environments. The Andrenam team brought 12 buoys fully ready for the exercise. We deployed six across two locations and successfully detected several unmanned underwater vehicles in real time. Mission success. More to come.

@occamsriver @paige_eden Who is Harvey, please don’t gate keep

@paige_eden Can I take a wild guess, are you talking about Harvey

Need to go back to the bodyworker who grabbed my abdomen and lifted up all my intestines and told me to tuck my pelvis to release the adhesion between my ascending colon and retroperitoneal fascia caused by pulling my psoas while cheerleading when I was 16 but he costs $560/2hr.

Solid-state batteries are changing naval warfare. And the threats they enable are far harder to detect. A post recently went viral claiming Iran's "Azhdar" UUV patrols at 18–25 knots with 4-day endurance and 600+ km range. This appears to be false, but the post went viral because the narrative is true. UUVs are a growing threat and maritime, as demonstrated by this crisis, is more important than ever. And the stakes are high. Iran has long held the Strait of Hormuz as a pressure point; roughly 20% of global oil supply flows through a 21-mile-wide chokepoint. UUVs capable of covertly laying mines in those waters would further threaten naval assets and escalate the conflict. UUVs are already being deployed by multiple nations. Today's lithium-powered vehicles cover hundreds of kilometers and patrol for days. Solid-state batteries at 400 Wh/kg, and climbing, mean that gets dramatically better, cheaper, and more scalable this decade. Sensing underwater is categorically harder than any other domain. cUAS has radars, cameras, RF, microwave. The electromagnetic spectrum works for you in air. Water kills RF. Acoustics are the primary detection modality, and littoral environments add clutter and complexity that open-ocean legacy systems were never built for. Those legacy systems were designed to find large submarines in deep water. Not swarms of small autonomous vehicles near ports, chokepoints, and critical infrastructure. As UUVs get cheaper, faster, and more autonomous - sensing is the defining gap. cUUV is going to be a massive market. Interdiction starts with detection. And detection underwater is still largely unsolved.