Sajith Wickramasekara
@sajithw
Followers
1K
Following
1K
Media
3
Statuses
13
San Francisco, CA
Joined June 2010
BoltzGen is coming to @benchling. We had a chance to sit down with the model co-creator @HannesStaerk to hear how BoltzGen was built, why transparent validation matters, and what sets this model apart. https://t.co/4OB2Axd9lC
benchling.com
Go "Behind the Model" with BoltzGen co-creator Hannes Stärk. Learn how Benchling and MIT built and validated BoltzGen for generative protein design, ensuring transparency and rigor in scientific AI.
Excited to release BoltzGen which brings SOTA folding performance to binder design! The best part of this project has been collaborating with many leading biologists who tested BoltzGen at an unprecedented scale, showing success on many novel targets and pushing its limits! 🧵..
0
5
36
Many scientists have told me @VerveTx is the best-run biotech they have ever worked at. I wanted to find out why. Very grateful to my friend @ambellinger for sitting down and sharing what made Verve special and lessons he learned as CSO. We covered: - Paying the pioneer's tax
3
2
9
Overnight success, a decade in the making. Congrats to the @uniQure team on data that bring real hope to Huntington’s patients! Grateful they were early Benchling adopters. Seeing firsthand how complex these medicines are to make inspired much of the tooling we’ve since built.
We announced positive topline data from the pivotal study of our #genetherapy AMT-130 for #huntingtonsdisease, which met its primary endpoint, demonstrating a 75 percent slowing of disease progression as measured by cUHDRS compared to an external control. https://t.co/h2ALjhjqTL
0
0
3
Come for @DrSynbio's inspirational journey building @StrandTx. Stay for the fun: sneaking posters into conferences and getting stumped by @geochurch during his thesis defense.
🎙️@DrSynbio shares the story of @StrandTx's first clinical trial patient — and the moment he heard the initial results. On #Transcribed, he and @sajithw talk about betting early on mRNA, why medicine is the best investment, and what biotech needs to deliver breakthroughs faster.
1
2
9
I've been working on this problem for over a decade. I've tried to distill what I've learned and articulate the two vectors by which AI can actually help us get more medicines.
1
2
58
Cloud computing helped bring science online, but it was only a prerequisite. The next wave of AI must be about systematization to increase speed and reduce costs.
1
0
19
Many industries evolved from craft to scalable. Toolmakers help. Synopsys made chip design modular, AWS turned data centers into a utility, Stripe unified payments.
1
0
20
Why? R&D today relies on artisanal workflows. When people hear artisanal, they think of craftsmanship, like a handcrafted watch, a carefully brewed espresso. But you also wait longer, pay more, and it’s scarce. Great for watches, not for drugs.
1
1
26
Over the last decade, FDA approvals have stayed flat, even as R&D costs rise. Science has made extraordinary leaps (CRISPR, mRNA, CAR-T, GLP-1s), but the systems around it still hold us back.
1
0
26
Everyone in biotech knows the numbers: over $2B and 10 years to bring a new drug to market. This gets repeated like it's a law of nature. But it's not. I’ve been thinking about it, and what that means for Benchling’s next chapter.
1
0
24
Long time lurker, breaking a decade of silence to share an essay on what I've learned from scientists about making medicines (spoiler: it's hard) and why the future looks bright (also spoiler: AI) 🧵
14
50
480