
SDE @JioTesseract | Humanoid Robotics | Simulations |@CERN'24 | XROS'23
Mumbai
Joined October 2021
- Tweets2K
- Following2.4K
- Followers318
- Likes2.6K
Pinned Tweet
Finally SDE at @JioTesseract
Grateful to everything!
馃搷 Reliance Corporate Park, Ghansoli, Mumbai
Saumy Sharma馃嚠馃嚦 retweeted
Teleoperating an OpenYAM using AR
It鈥檚 very satisfying to see the overlay match the physical motion.
Hand tracking on @specs feeds the IK solver from @dimensionalos which drives the physical arm.
Saumy Sharma馃嚠馃嚦 retweeted
We just killed Notion, and open-sourced it.
Introducing Terminus Notes: everything you like in Notion, removed the stupid AI credits and confusing sidebars.
100% Free, 100% Open source, 100% Customizable. You can export your data or point it to your Claude at any time.
Saumy Sharma馃嚠馃嚦 retweeted
We finally put together a video showing what we鈥檝e been building at Preload (YC F26) 馃馃
We collect synchronized egocentric video + EMG + tactile data for robot manipulation, VLAs, and world models.
If you need robotics data, we can collect it end to end, at scale, starting as soon as tomorrow, and help integrate it into model training.
We鈥檙e also excited to work with companies exploring where dexterous manipulation can unlock new automation.
馃帴 Launch video below
preload.ai
[email protected]
i found the exact 6-month path to become a robotics engineer
everyone on your timeline is rushing into AI engineering right now.
and hitting the same wall - it's crowded and a subagent can do most of it for pennies.
here's the roadmap, zero electronics knowledge to a working robot arm:
/ month 1-2: ohm's law, arduino, motors - your first robot for under $100;
/month 3: CAD your own parts, 3d print them, build the SO-101 arm;
/ month 4: ROS 2, gazebo simulation, SLAM, autonomous navigation;
/ month 5-6: the control theory and robot learning that makes it click.
and you get:
/ a github full of robots that moved not courses you finished;
/ 100+ verified resources and prices, checked september 2026;
/ a $122 open-source arm you can teleoperate and train from demonstrations.
you basically become the one engineer who can make a motor turn AND train a policy - most people can only do one.
P.S. the full guide runs almost 12000 words across 27 practical builds.
In job we are using AI and at home it's all Instagram looks like our generation is just a revenue model for tech companies we spend major time either with Claude, chatgpt, gemini and call it productive on the other hand for entertainment it's just Instagram.
1
146
Finally got my @cyberwave_com swags
Thank you for the support and being a helpful community hoping to make an impactful project with @Vittoriobanfi @khushiSharma_22
but now the designated contact has stopped responding for days.
The insurance company requires the original/offline documents, and this delay is putting our claim at risk.
My father spent a few days at Patanjali Yoggram, and the treatment was genuinely helpful. I appreciate that
1
98
I鈥檝e been trying to get his physical medical documents couriered because I can鈥檛 travel there. Initially, the staff was very supportive, but now the designated contact has stopped responding for days.
A simple response or dispatch update would go a long way @Yoggramofficial
56
Saumy Sharma馃嚠馃嚦 retweeted
you don鈥檛 need a robot to see physical ai in action, you just need a browser.
this week I finally tried something I've wanted to do for a while: design an SO-101 scene in three.js, connect it to a real leader arm, collect teleoperation data, train an ACT model, then port it to run in the browser.
the result is the first ever demo of ACT that runs locally on the web.
try it yourself now at lerobot.binhph.am
Saumy Sharma馃嚠馃嚦 retweeted
This is Anagha Rajesh.
And she is crazy.
She wants to store data in bacterial DNA.
Her startup is called BioCompute.
And last year, they actually did it.
They stored data in DNA and retrieved it in their tiny lab in Bengaluru.
This is a huge moment.
But why is she even doing it?
Because DNA is the ultimate storage tool available to us.
Just 1 gm of DNA can store about 215 petabytes of data - that's like storing over 2 million movies in 4K.
On top of that - this data can last for literally 1000s of years.
Right now, they still need to figure out a way to make the reading and writing process faster and cheaper.
But if BioCompute solves this problem - we could theoretically store all the data created in the world every year in the palm of our hands.
And that would be insane.
P.S. Check out this video from @vy0mbhatia going to Anagha's lab and actually doing it.
why did we put chaar diwari inside of a bacterium's dna?
because a silicon chip can only fit transistors till the size of atoms. however, a single gram of DNA is so dense that theoretically it can store over 200 petabytes of data.
DNA is so durable that info on it can survive for hundreds of years at room temperature.
tldr; DNA can be silicon chips and even better. and biocompute is early.
$3500 per MB, is a cost that most DNA companies put it. mainly because DNA is expensive to make from scratch.
but a chemical engineer from BITS Pilani thought what if there's no making it from scratch and we just pick it out of abundant bacteria?
at 23, Anagha is chasing the impossible. and all this is happening in a small lab in 1st Block Koramangla, Bengaluru.
her cost? $1 per MB.
a person who deeply cares. that's what india needs.
but she wants to go further, $1 per TB of storage. which is theoretically possible with a 'micro-fluid' drive. beating the cost of existing storage solutions at $5-6 per TB.
and hence, to show this one of a kind journey, chaar diwari is now immortalised inside of a bacteria's dna.

















