Samarth Jajoo (@jajoosam) header
Samarth Jajoo (@jajoosam) profile picture

making therapeutics

Hyderabad, India
Joined December 2014
Photo by Samarth Jajoo (@jajoosam)Photo by Samarth Jajoo (@jajoosam)Photo by Samarth Jajoo (@jajoosam)Photo by Samarth Jajoo (@jajoosam)Photo by Samarth Jajoo (@jajoosam)Photo by Samarth Jajoo (@jajoosam)Photo by Samarth Jajoo (@jajoosam)Photo by Samarth Jajoo (@jajoosam)Photo by Samarth Jajoo (@jajoosam)Photo by Samarth Jajoo (@jajoosam)
There's so much we don't know about cancer vaccines -- excited to make our first one in the open :)
We’re making a personalized cancer vaccine for Yash, a 34 year-old pancreatic cancer patient, in the open with his consent. We want external experts to review our design and make their own contributions before we manufacture it at the RNA Foundry. RNAFoundry.com/yash
2
16
117
4,259
Samarth Jajoo retweeted
Beware of those who peddle more bureaucracy as a solution to the problems of Indian biotech. More consortia, panels & committees aren't going to solve anything -- we need policy changes & streamlined funding to empower individual institutions, companies, and scientists to do R&D! Some simple first steps from the startup side: -- Faster imports with lower duties for critical reagents and equipment; establish a fast track like China has -- Removal of burdensome paperwork and approvals (including the need for a physical PAN card) that make it nearly impossible for foreign investors to put small cheques into Indian-incorporated companies -- Government funding for startups of INR 10-50 crore per R&D project without the need for matching funds or payback; the new RDI Fund requires at least 50% matching, which small startups typically cannot provide -- Faster Phase I clinical trial approvals for innovative new medicines, and the establishment of rapid compassionate-use and named-patient pathways to open up access to experimental medicines for patients who have exhausted existing treatment options
13
36
2
203
7,525
what would establish that a simulated brain is a ‘real’ emulation? the recent viral ‘fly playing beat saber’ videos has made this question surprisingly prescient. @davyneuro & I argue that whole-brain emulation needs a shared benchmark & propose what it should test! >>
17
11
2
119
6,768
Samarth Jajoo retweeted
type shit -Tolkien
Tolkien used 'cooked' in precisely the modern sense.
1
64
6,934
We've spent the last 30 months building the physical infrastructure to develop personalized medicines. Now, we deploy: in the next year, we will manufacture 100+ mRNA cancer vaccines - each purpose built for a single person. RNAFoundry.com
Launching RNAFoundry.com – we manufacture personalized cancer vaccines.
17
55
5
399
15,456
It may be possible to train music generation models which have a shot at treating, and perhaps even preventing Alzheimer's. It'd take ~$5m to build and run trials for what could become one of the most widely deployed public health interventions. jajoo.sh/gamma
5
15
5
194
11,455
Samarth Jajoo retweeted
I wonder if an Indian deeptech fund run by actual deeptech founders could raise $100m... @akadogluk @Itamar_Bitan and I have been talking about this. Maybe @ashleevance can find us some LPs. We should call it the Cowardice Fund! infinitesunrise.com/p/overco…
Why are Indian VCs so afraid to take risk and make money? Someone should go to India and take risk and make money. Our story - corememory.com/p/a-new-india…
10
8
2
93
6,259
The team: - computationally designed and tested ~1300 vaccine constructs in animals - screened ~5000 novel LNP formulations in the wet lab - built a mRNA-LNP manufacturing facility which can produce a million doses every year and now, we are well into our first human trial!
On August 31, the first two participants were injected with @PopVaxIndia’s first fridge-stable RNA vaccine candidate in a Phase I clinical trial in Australia. We built this vaccine at our RNA Foundry in Hyderabad – from computational design to GMP manufacturing. Here's how:
14
10
1
133
5,433
as far as building novel molecules goes, this was non-trivial on every end. we had to build out a ton of infrastructure from scratch. we now possess the muscle to move from design to the clinic at what I expect will be unprecedented pace and effectiveness. more soon!
11
500
Samarth Jajoo retweeted
I am hiring a talented and tasteful full-stack filmmaker for a three-month fellowship @PopVaxIndia, full-time in Hyderabad, to help the world imagine what the future of medicine will look like in our new world of abundant intelligence. We think medicine will be increasingly personalized and preventative, and the mechanics of healthcare will have to be entirely transformed to make that possible. A lot of this will look more like making science fiction than producing launch videos, though there will be some of that as well. I'm looking for someone who is as comfortable shooting and editing a short film entirely in real life as they are generating it entirely with AI, with the ability to seamlessly embed the one medium in the other. If this is you, email me at film [at] popvax dot com with a few samples, ideally both writing and video. This is a paid position, and it has the potential (but not the requirement) to convert to a long-term role if there is mutual alignment. My brother @SahajSankaran and I write and direct a little on the side -- see youtube.com/@honestyontape -- so, for better or worse, I will be quite opinionated about what we put out. If this goes well, it will be a deep and fulfilling collaboration on all aspects of the creative process, from the writing to the final cut. I should have done this a long time ago (many such cases). This will be a lot of fun! @pbshgthm @ashleevance @oshosidhant @anupamsidhant @pavlikam @ABiggerSpalash @neervj @neilshroff @sidin
5
25
6
89
11,056
Samarth Jajoo retweeted
There’s lots of buzz around agentic harnesses for robots, particularly for long horizon tasks that require complex reasoning and memory. But what will it really take to turn a reasoning agent into a reactive, reliable and low-latency robot policy? In our new paper, workspace models, we design a new memory architecture that acts as a latent harness for stronger reasoning models. Here’s why we think this might be the way forward: (1/9)
8
58
15
214
35,947
Samarth Jajoo retweeted
AI allows us to design & develop life-saving medicines in months, yet India takes a year to approve clinical trials – Australia & China take weeks. Honoured to co-write an op-ed in @EconomicTimes with @KiranShaw on decentralising our trial framework to accelerate Indian biotech.
12
31
4
154
8,385
While Yun is in the news for yet again coming up with clever ways to do variant effect prediction, I want to remind folks about a really neat result from his lab ~4 years ago: biorxiv.org/content/10.1101/… Protein language models learn multiple conformations!
2
1
78
7,270
How do we disentangle them? I think a method that figures out a way to do that is our best shot at fast progress in generating multiple conformations / modelling conformational diversity in proteins. @yun_s_song curious if you've thought about this more!
1
496
Amazing! I think the technology Anand and Melody create will help illuminate the mechanism behind pathologies and ultimately lead to diagnostics + therapeutics we otherwise wouldn't even be able to imagine.
Hello world! We’re @MelodyFRO. Most wearables don’t go far beyond heart rate and activity. Lab diagnostics paint a deeper picture but don’t read continuously. Melody is building biosensors, devices, and standards to bridge this gap. Hear our story at melodyfro.org
1
11
2,063
Samarth Jajoo retweeted
Excited to share our work on a model for mapping molecules to olfactory receptors to perception, out today! Check out the link in thread ⬇️
Out today in Cell Systems: the final version of our work on biologically-inspired models for predicting how molecules smell! With @SeyoneC and @jdthamores
6
16
4
84
8,722
Incredible work, and another win for inference time optimization!
We introduce ConforNets, a mechanism for conformational control in AlphaFold3 models - SoTA at producing diverse conformations on every multistate benchmark (N=104) - Novel capability: transfer state from one protein to another Outperforms BioEmu, ConforMix and AFsample3 🧵1/8
4
65
6,161
Samarth Jajoo retweeted
My wonderful brother @SahajSankaran and I are pleased to release The Project, a new comedy pilot we wrote and directed about a group of villagers who start a fake consulting firm in a last-ditch attempt to save their village from demolition. Take a look!
(THIS IS SATIRE) Progress has costs. Sometimes, those costs are jobs. Sometimes, they are people. Sometimes, it can be whole villages also. What to do? As the village of Chanala learns, the bulldozer of development cannot be stopped, certainly not by lying in front of it. (1/3)
1
6
1
46
4,561