Discover and develop novel materials with the world's most advanced self-driving lab.
Joined December 2023
- Tweets 203
- Following 2
- Followers 3,715
- Likes 35
96 Photos and videos
Pinned Tweet
2 Dec 2025
History is defined by the materials we master. But the materials of the past cannot build the future. Radical AI is building scientific intelligence to autonomously discover and manufacture next-generation materials to solve the world's greatest challenges.
12
63
323
311,370
Jun 12
AI's hardest unsolved problems lie in the physical world.
We showcased our agentic scientist last week at Critical Mass for NYC Tech Week.
Thank you @Newlab for having us and bringing together the teams developing critical technologies in NYC.
3
8
623
Jun 11
To invent at the speed of AI, we must make materials science data actionable.
Traditionally scientists are limited by scattered data, manual workflows, and inconsistent analysis.
Our latest automated workflow captures structured images at scale, so every sample is directly comparable.
What once took hours of manual measurements now takes roughly a second.
4
2
11
829
Jun 11
Read more about what we built to analyze microstructure at scale: radical-ai.com/news/scale-sp…
1
175
Jun 10
Building a self-driving lab means automating annealing.
Until recently, determining annealing time required measurements taken by hand. Our system can interpret alloy SEM microstructure at scale, and develop a clean dataset of structured data and standardized images for inventing materials of the future faster.
To our knowledge, no one has done this before.
1
9
3,642
Jun 10
Read more about what we built to analyze microstructure at scale: radical-ai.com/news/scale-sp…
1
3
204
Jun 9
The experimental data needed to discover new materials doesn't exist.
Our work started where that data is made: in the lab.
Closing the loop with a self-driving system is how to invent materials that make it to industry.
1
2
6
675
Jun 4
What worked for biotech doesn’t transfer to material science.
Material invention won’t come from simulation, it will happen in the lab.
1
4
12
880
Jun 2
There's a difference between creating value and capturing it.
Radical AI is constantly pressure testing the path toward building the world's largest experimental dataset, and not afraid to take the risks to there.
1
6
413
Jun 1
"One recent campaign discovered around 300 novel compositions over 16 weeks," reports @adele_peters for @FastCompany.
Created in our-closed, loop self-driving lab, these novel materials outperformed current industry-leading materials.
1
3
433
May 28
The world doesn't move forward if no one succeeds here.
Not everyone gets it.
1
9
641
May 28
“This AI-driven process is really a radical shift to allowing us to scale scientific discovery,” says @josephfkrause. “You go from 10 scientists focused on one problem to one scientist focused on 10 problems at a time."
Thank you @adele_peters @fastcompany for the profile on our first of its kind active learning loop lab.
1
7
15
1,765
May 25
In software, a unit test takes seconds. In materials science, an experimental process takes weeks or months.
@josephfkrause on why real-world experimental data, including the negative results no one captures today, creates an immense moat when combined with AI.
1
2
393
May 22
Traditional materials science is serial: hypothesize, simulate, synthesize, analyze, repeat. Each step waits on the last.
@josephfkrause on how Radical's robotic self-driving lab inverts that, running hypothesis, simulation, experimentation, and analysis in parallel, and learning from all of it simultaneously.
1
1
9
634