Pinned Tweet
I'm Amit. Anesthesiologist in NYC. I build things in healthcare, AI, and fitness.
YC S17. Currently: Pixie, StatPhone, ProofLeague.
amithpatel.com
🤖 Made with AI
As a physician who bikes in NYC, I’m grateful for our bike lanes. But too often they’re blocked, forcing cyclists into traffic, while nearby police do nothing. @NYCMayor, can we make keeping bike lanes clear a consistent enforcement priority?
Opus 5.5 is incredible. Single shotted this reel for my free app - ProofLeague (ProofLeague.com)
Such a promising project!
I've attempted to map everything in oncology in a public website and open source repo for all.
The site is trying to get all cancers, products, technologies, bottlenecks, people, startup opportunities with 1000+ ideas for upgrading the field.
If you have a loved one with cancer and you are technical or can engineer, go take a look, file improvement requests or bugs or help with the open repo and make this the best open and free info resource for individuals, researchers and educational use.
This should save people time, aid AI oncology projects and generate positive action. If you are not technical just complain in this thread about broken or annoying or things you want and I'll fix them live.
Some of the interesting pages:
Treatments: onco.cc/drugs
A gallery of the molecules being used onco.cc/molecules/
And targets: onco.cc/targets/
The technologies in oncology: onco.cc/technologies/
1100 ideas for helping oncology: onco.cc/ideas/
Bottlenecks on oncology: onco.cc/bottlenecks/
Open questions (LETS GO RESEARCH PEOPLE) onco.cc/open-questions/
Startup requests (LETS GO STARTUP PEOPLE) onco.cc/startup-requests/
Mechanics of cancer: onco.cc/mechanics
Battlefronts: onco.cc/fronts/
Isotope supply: onco.cc/isotopes
Key papers: onco.cc/key-papers
Pipeline funnels: onco.cc/pipeline
Cancer by type: onco.cc/cancers/
Institutional rankings: onco.cc/institutions/
The startups: onco.cc/startups/
Heros and heroines : onco.cc/heroes/
Key medical people: onco.cc/people/
Here is the project roadmap: onco.cc/roadmap/
Models and data sets: onco.cc/models/
There are other views as well, take a browse.
Try making a PR if you have an upgrade to this on the repo here: github.com/judegomila/OnCo
If you are biologically/medically minded and something is wrong, file a bug as well or say on the thread and we will get it fixed live.
If this is a useful project star the repo and help get it calibrated.
I've tried to add some other languages but I cannot speak them so tell me if that doesnt work well.
I believe we will crack oncology and having total information dominance is key to the problem.
Let the feedback flow!
Amit Patel, MD retweeted
I've attempted to map everything in oncology in a public website and open source repo for all.
The site is trying to get all cancers, products, technologies, bottlenecks, people, startup opportunities with 1000+ ideas for upgrading the field.
If you have a loved one with cancer and you are technical or can engineer, go take a look, file improvement requests or bugs or help with the open repo and make this the best open and free info resource for individuals, researchers and educational use.
This should save people time, aid AI oncology projects and generate positive action. If you are not technical just complain in this thread about broken or annoying or things you want and I'll fix them live.
Some of the interesting pages:
Treatments: onco.cc/drugs
A gallery of the molecules being used onco.cc/molecules/
And targets: onco.cc/targets/
The technologies in oncology: onco.cc/technologies/
1100 ideas for helping oncology: onco.cc/ideas/
Bottlenecks on oncology: onco.cc/bottlenecks/
Open questions (LETS GO RESEARCH PEOPLE) onco.cc/open-questions/
Startup requests (LETS GO STARTUP PEOPLE) onco.cc/startup-requests/
Mechanics of cancer: onco.cc/mechanics
Battlefronts: onco.cc/fronts/
Isotope supply: onco.cc/isotopes
Key papers: onco.cc/key-papers
Pipeline funnels: onco.cc/pipeline
Cancer by type: onco.cc/cancers/
Institutional rankings: onco.cc/institutions/
The startups: onco.cc/startups/
Heros and heroines : onco.cc/heroes/
Key medical people: onco.cc/people/
Here is the project roadmap: onco.cc/roadmap/
Models and data sets: onco.cc/models/
There are other views as well, take a browse.
Try making a PR if you have an upgrade to this on the repo here: github.com/judegomila/OnCo
If you are biologically/medically minded and something is wrong, file a bug as well or say on the thread and we will get it fixed live.
If this is a useful project star the repo and help get it calibrated.
I've tried to add some other languages but I cannot speak them so tell me if that doesnt work well.
I believe we will crack oncology and having total information dominance is key to the problem.
Let the feedback flow!
Amit Patel, MD retweeted
I'd love to see what korea has in mind for oncology and what technologies we can distribute.
Where is the samsung of cancer?
☕️ 이건 정말 의미 있는 프로젝트라, 한국어로도 간단히 소개하고 싶습니다.
@judegomila가 암 관련 자료를 `onco.cc`에 정리해두었습니다.
암종, 치료제, 타깃, 임상시험, 기술 경로, 기업, 연구기관은 물론 오랫동안 풀리지 않은 문제와 연구 공백까지 사이트 안의 관계 그래프를 따라가며 살펴볼 수 있습니다.
프로젝트는 GitHub에도 오픈소스로 공개돼 있습니다.
자료를 직접 읽을 수 있고, API·CLI·MCP를 통해 AI가 검색과 정리를 보조하도록 연결할 수도 있습니다.
의료 AI, 연구 도구, 데이터, 신약 개발을 하는 분이라면 특히 `Gaps`와 `Bottlenecks`를 한 번 봐주세요.
지금 만들고 있는 것이 누군가 놓친 한 조각을 채울 수 있을지도 모릅니다.
신뢰할 수 있는 자료 하나를 보태고, 오래된 정보를 바로잡고, 유용한 Issue 하나를 남기는 일도 암 정복을 향한 의학의 작은 전진이 될 수 있습니다. 🧬
🔗 onco.cc
🔗Github github.com/judegomila/OnCo
#AI #AICaffeine #Github
Amit Patel, MD retweeted
A 9/11 widow just rent rogue in NYC…
She read her husbands name and then called out Islamic terror and Saudi Arabia.
She called out Bush, Obama, Biden and JD Vance/Trump while they sat right in front of her. The entire crowd at ground zero cheered.
Apple is drowning in vibe-coded apps and review times blew out to a month. Shipping got cheap. Judgment did not.
Cure most diseases in 5 to 10 years is a hell of a claim. I will take AI that just does not miss the anticoagulant.
This quoted post is unavailable.
Been running for a while and I still got PF syndrome. Going to explain it because half of what I see about knee pain is wrong.
You don't have to stop running. You have to stop doing the version that keeps poking it. Easy flat miles if the pain stays reasonable. Skip the speed and the hills for a bit.
The demo is easy. Converting to patients is the part most healthcare AI never shows.
an AI agent is running facebook ads for a local business
it just got the 2170 ebook downloads in august
16% of people who download the ebook turn into a future patient
they are healthcare business
so they got 347 new patients
average customer lifetime value is $2500
so $5600 in ad spend turned into... you do the math
the funnel is facebook ads > ebook download > quiz funnel > email nurture
so
How the agent works:
- researched pain points and desired outcomes of target customer
- make creative promoted the ebook based on this resarch
- uploads creative through the API
- agent media buys the account
- learns from best performing ads
- makes more ads like the best performing
if you want this for your local business get it below
Color coding the cab so you don't climb into the wrong one. That's a real problem if these things actually show up at the airport.
Cybercab RGB Light Bar in use while performing real world pick-ups and drop-offs in downtown Austin, TX!
🎥 @artsimage with @AdanGuajardo
London ran AI on a live pituitary case. Not a scan from last week. The model watched the video and marked the vessels while they took the tumor out. That's the version I actually want in a room.
I use Grok to read medical papers. Not instead of reading them. So I don't get hypnotized by the abstract.
Amit Patel, MD retweeted
Primary care is the lowest-paid physician specialty.
Wanna see what my day looked like today?
And somehow I’m constantly doing specialist work because patients can’t afford the specialist, don’t want to go, or literally can’t get an appointment for months.
People on Twitter keep telling me AI is going to automate primary care.
Good luck. I'm more focused on how to get paid more appropriately for all the things that I'm doing.
Here was my day today. Nothing unusual. Just an average day in primary care:
1. High cholesterol follow-up + weight-loss medication management.
2. “Annual physical.” Found a suspicious skin lesion concerning for cancer. Also noticed word-finding difficulty → started a dementia workup.
3. Hospital follow-up after a recent heart attack complicated by cardiogenic shock.
4. New dysphagia + anxiety → started workup.
5. “Annual physical.” Also menopause + hormone replacement therapy management.
6. “Annual physical.” Also follow-up after discharge from an inpatient psychiatric hospitalization.
7. 72-year-old with unintentional weight loss + coughing up blood. Concern for a new cancer → started the workup.
8. “Annual physical.” Also severe agoraphobia. He has essentially isolated himself on his property in a bunker for two years. We talked about how to start getting him back into society.
9. Annual physical. An actual annual physical!
10. “Annual physical.” Also managing multiple side effects from chemotherapy.
11. “Annual physical.” Found another suspicious skin lesion. Also started weight-loss medication and hypertension treatment.
12. “Annual physical.” Also started treatment for alcohol use disorder and connected the patient with resources.
13. “Annual physical.” Also discussed pregnancy, a recent miscarriage, and the anxiety and stress surrounding it.
14. Annual physical for a healthy guy who wanted to nerd out with me over additional biomarkers.
15. Psychiatric medication issue because the patient can’t get into psychiatry. Concern for a manic/bipolar episode + lab follow-up.
16. “Annual physical.” Also anxiety treatment, weight-loss medication management, and hypertension treatment.
17. Pulmonary fibrosis with worsening shortness of breath and anxiety. Discussed end-of-life wishes. Signed an out-of-hospital DNR and arranged additional support.
18. STI visit and treatment.
19. Type 1 diabetes + insulin pump management because the patient can’t get into endocrinology. So now primary care is managing it.
20. Virtual visit for a newly diagnosed Parkinson’s patient. Starting treatment because neurology is booked out for the next NINE MONTHS.
That was one day.
Dermatology. Cardiology. Psychiatry. Oncology. Endocrinology. Neurology. Women’s health. Addiction medicine. Palliative care.
And buried underneath all of it: prevention, hypertension, cholesterol, obesity, diabetes, anxiety, medications, labs, and the actual annual physicals.
Primary care increasingly feels like the specialty that absorbs whatever the rest of the healthcare system cannot handle.
We’re paid less because what we do is labeled “primary.”
There’s nothing primary about it.
So yes, AI will absolutely change my job. I hope it does. There is plenty of administrative garbage I’d love to automate.
But replacing the person who has to walk into 20 completely different rooms, figure out what actually matters, recognize what doesn't fit, know when something is dangerous, manage what the patient can't get managed elsewhere, and then immediately reset for the next completely unrelated problem?
Good luck automating that day. Yes some of it can be automated ..but the majority..not anytime soon.















