sanjayjpatel

CEO @ Mediflix | Building AI-powered healthcare search | Angel investor | Dad, tennis player, curious about longevity + robotics

[sanjayjpatel @sanjayjpatel]

We’re building AI-powered healthcare search that solves for trust, literacy, and accessibility. Healthcare doesn’t just need more content, it needs better ways to find and use the right content.

Sharing thoughts on AI, healthcare, startups, investing, tennis, and family life.

[sanjayjpatel @sanjayjpatel]

@MartinGTobias Essentialism, great book

[Sheel Mohnot @pitdesi]

Alfred Lin's advice to startups

[Sheel Mohnot @pitdesi]

David Marcus, former president of PayPal has a great take on what happened- PayPal stopped being product led and was financially led (poorly!)

https://t.co/0BbcMRdB6f

[Morgan Cheatham, MD @morgancheatham]

after a week of conversations at JPM, the patterns are clear, and they align with our thesis. capital, talent, and economics are rapidly reorganizing around three key axes: computation, precision, and prevention.

next week, we're releasing the Breyer Capital 2026 Healthcare & Life Sciences Predictions—ten inflection points where technological maturity meets market readiness, featuring perspectives from our top portfolio companies.

going into the long weekend, here are three predictions we feel most strongly about:

[sanjayjpatel @sanjayjpatel]

@bhalligan I was surprised went to a Boston AI event three weeks ago and didn’t meet a single investor. I think if we bring the investment talent, the rest will follow.

[sanjayjpatel @sanjayjpatel]

@elonmusk @CathieDWood Within four years is key here. Over time smartphone will win in terms of cost.

[sanjayjpatel @sanjayjpatel]

@guilleflorvs FOUNDER

[sanjayjpatel @sanjayjpatel]

@SahilBloom Can get leverage in stocks too, can someone take that and also 2 percent annual in property costs and maintenance into account?

[sanjayjpatel @sanjayjpatel]

@MattOrlic Aware

[sanjayjpatel @sanjayjpatel]

@monicaa_AI Leads

[sanjayjpatel @sanjayjpatel]

@thogge Would like to introduce you to a vertical fintech for athletes

[sanjayjpatel @sanjayjpatel]

@bryce Vertical Fintech for athletes and also vertical healthcare search

[sanjayjpatel @sanjayjpatel]

@henrythe9ths Fundraise

[Bill Gurley @bgurley]

I have written my first book! A passion project of almost 10 years, Runnin' Down a Dream aims to give people both the motivation & the methods for thriving in a career they actually love. Put a lot of heart and soul into this - hope you ❤️ it. Pre-order: https://t.co/vR4XG6JxDy

[Mark Hyman, M.D. @drmarkhyman]

Did you know Parkinson's disease is largely preventable?

Two of the world's leading neurologists, @RayDorseyMD and @MichaelOkun, just laid out the evidence, and it changes everything we thought we knew about this disease. 👇

[sanjayjpatel @sanjayjpatel]

@mhp_guy That’s what I do, actually goes faster because I can make decisions about materials and floor plan on the spot.

[sanjayjpatel @sanjayjpatel]

AI has shifted what’s sets companies apart. In tech it was always “invest in the product”, now its “is the company ready for go to market”.

[Michael Okun @MichaelOkun]

Not so fast: Why stick w/ your human doctor over AI? In the rush to embrace artificial intelligence in medicine, a just published study by Nori and colleagues in Computation and Language introduced a powerful AI system that 'outperformed physicians' in tough diagnostic challenges. However, before we hand over the stethoscope, let’s consider what this means for real-world care.
Key Points:
- AI outperformed physicians on complex NEJM CPC cases using a new system (MAI-DxO).
- AI had an 85.5% diagnostic accuracy.
- AI was 4X better than average generalist doctors.
- MAI-DxO reduced testing costs by 20–70% compared to physicians and to standard AI models.
- The authors employed a virtual doctor panel to more efficiently reason on cases.
- The authors introduced a dynamic, step-by-step diagnostic benchmark more closely mimicking real-world clinical reasoning.
- The author chose not to use more static multiple choice type tests.
My take: There were 5 points that resonated w/ me, and also left me wondering if the study was too biased against the human element. 1- AI is getting smarter, but it's not ready to replace your doctor. This study showed AI could reason through complex cases. Keep in mind these were the hardest, and the cases heavily curated. Everyday medicine is 'messier.' 2- Humans bring context and compassion. AI lacks bedside manner, judgment in uncertainty and the ability to factor in your emotions, values and life circumstances. 3- Even 'superhuman performance' has blind spots. The study didn’t include common cases, benign conditions or the messiness of real clinics. AI doesn't account for screaming toddlers, time limits and insurance hurdles. 4- Your doctor isn’t just a diagnostician, they’re your guide. Physicians know when to worry, when to wait and when to hug a patient, instead of simply ordering another test. 5- The best care will come from collaboration, not replacement. The future of medicine will be human expertise 'enhanced by AI,' not overshadowed by it. I say YES to getting excited about the promise of AI, but I also say 'stay grounded' in what truly heals. AI will not replace human connection, listening and the art of medicine. Remember, complex cases take time to unravel and it is a shared journey. Doctors should be assisted by AI, but not replaced.
https://t.co/Uak3u2ZDep @ParkinsonDotOrg @FixelInstitute #AI #parkinson #Alzheimers @NEJM

[Aaron Levie @levie]

The reason why going AI-first as a company is so important now is best captured by Paul Graham's startup advice of "live in the future, then build what's missing." AI-first companies will see new problems to solve soonest, and that flywheel will only compound over time.