Filip Tosh

[Filip Tosh @Filip_Tosh]

instead of doom scrolling read this masterclass on how to become an AI engineer in 2026

it’s never been easier.

[Rohit @rohit4verse]

https://t.co/3MgR2Mp5AV

[Filip Tosh @Filip_Tosh]

@TheEchoOS Lets gooo

[Filip Tosh @Filip_Tosh]

desci + ml-powered feedback loops + permissionless experimentation + FDA support

the future of drug development has arrived and its on chain

[Dr. Marty Makary @DrMakaryFDA]

FDA is now open to Bayesian statistical approaches. A leap forward!

Bayesian statistics can help:
✅ Clinical trial design
✅ Finding the optimal dose
✅ Extrapolation to children
✅ Leveraging phase 2 results in phase 3

[Filip Tosh @Filip_Tosh]

we have enough ideas for peptides

the bottleneck is better systems for testing more compounds, faster, and learning which ones actually achieve desired effects

[Filip Tosh @Filip_Tosh]

what if there was a better and faster way to compound the lessons learned from micro experiments across science

[Filip Tosh @Filip_Tosh]

drug development is fundamentally an uncertainty reduction problem.

centralized funding forces early irreversible bets.

desci spreads risk across parallel experiments and lets capital follow what actually works.

thats why desci wins.

[Filip Tosh @Filip_Tosh]

@SMER28RAPA desci is the future of drug development.

results look promising. keep building

[Filip Tosh @Filip_Tosh]

desci and clinical trials are about to compound learning the way software did in the 2000s.

[Filip Tosh @Filip_Tosh]

2026 is the beginning of the shift to a decentralized pharma industry

1. biology is too complex for single institutions to explore alone.
2. centralized trials learn too slowly for modern disease timelines.
3. innovation compounds only when experimentation isn’t permissioned.

that last one is the most important. those of us building on decentralized rails will soon be publishing discoveries too significant for trad institutions to ignore.

[Filip Tosh @Filip_Tosh]

science is for everyone.

[Filip Tosh @Filip_Tosh]

@elianoayounes 11/10

[Filip Tosh @Filip_Tosh]

been heads down building for months.

time to show some work.

building the data infrastructure layer for modern drug development.

more soon.

[Filip Tosh @Filip_Tosh]

@elianoayounes 😮‍💨

[Filip Tosh @Filip_Tosh]

@elianoayounes oooo

[Filip Tosh @Filip_Tosh]

@elianoayounes congrats!

[Filip Tosh @Filip_Tosh]

one of the most relevant posts i’ve seen on x

[Ben Tossell @bentossell]

https://t.co/Ref8GgkIR5

[Filip Tosh @Filip_Tosh]

GM

who else is building today?

[Filip Tosh @Filip_Tosh]

(a+b)² = a² + 2ab + b²

(a-b)² = a² - 2ab + b²

(a+b+c)² = a² + b² + c² + 2ab + 2bc + 2ca

(a+b)³ = a³ + 3a²b + 3ab² + b³

(a+b)³ = a³ + b³ + 3ab(a+b)

(a-b)³ = a³ - 3a²b + 3ab² - b³

(a-b)³ = a³ - b³ - 3ab(a-b)

(a+b+c)³ = a³ + b³ + c³ + 3(a+b)(b+c)(c+a)

a³ + b³ = (a+b)(a² - ab + b²)

a³ - b³ = (a-b)(a² + ab + b²)

a³ + b³ + c³ - 3abc = (a+b+c)(a² + b² + c² - ab - bc - ca)

[Filip Tosh @Filip_Tosh]

to my founder friends: use inexperience to your advantage

uncertainty breeds experimentation which is where true growth is born

twenty years of experience often teaches you what's "supposed" to work

sometimes not knowing the rules lets you see what everyone else misses

[Filip Tosh @Filip_Tosh]

@elianoayounes best merch