Abhiraj

loving the problems between biology and engineering

currently @ScienceCorp_, prev eng @AbbottCardio, prev founder @WearZeno, student @CoulterBME

[Abhiraj @abhididthat]

“But if we can internalize the statistical strangeness of our current selves, at least the present will gain some of the credit it surely deserves”

[Abhiraj @abhididthat]

This is awesome
https://t.co/WkiboKlCSC

[Abhiraj @abhididthat]

@sahilchat YW. Js a journal of thoughts

[Abhiraj @abhididthat]

Animation really used to take crazy effort. The earlier Arthur seasons used to take 6 months to make an episode.

Sidenote: I think watching Arthur as a kid made a significant positive impact on me. PBS Kids was awesome

[dylan ツ @demian_ai]

fascinating that they managed to make a completely different medicine for every single patient, and get it through a large phase 3 trial.

Here’s what that actually required:
1. after surgery, they sequence the tumor.
2. an algorithm then ranks the mutations that make that specific cancer look foreign to the immune system and picks the best
3. hose get written into a personalized mRNA vaccine.
4. separate system schedules and manufactures each unique batch so it arrives on time.

The ranking rules were locked in place. No continuous learning mid trial, just a fixed, high stakes decision system that had to perform the same way for over a thousand patients under regulatory scrutiny.

a constrained system that turns a biopsy into a one of a kind medicine fast enough and reliably enough to matter clinically.

If this approach starts working in other cancers, the constraint shifts dramatically:
-> every new patient becomes a separate, regulated compute job: sequence analysis, mutation ranking, vaccine design, one batch manufacturing

The demand moves to:
- more training power.
- accurate, auditable systems that can run the same high stakes loop over and over (with hard deadlines)

That’s a different kind of AI problem than most of the industry is currently optimising for, and probably one of the best ones to solve.

[Abhiraj @abhididthat]

On seeing faces in cars:

Is it more that we find a way to look for human in everything so we see the front as a face?

Or is that from the beginning of car design, designers had the innate feel to add a touch of human beauty in cars?

[Abhiraj @abhididthat]

Why is there a Model Y at monterey motorsports festival? I’m thinking it’s gotta be a joke? Not a tesla hater but this is just wild😂

[Abhiraj @abhididthat]

Holy shit i’m in love

[Abhiraj @abhididthat]

Constraint breeds creativity

[Abhiraj @abhididthat]

Remember when teachers would use these #legendary

[Abhiraj @abhididthat]

Literally

[Abhiraj @abhididthat]

How the crowd reacted when Sam walked in was like hearing Warriors fans cheer for Curry’s lineup introduction. Engineer fanbases are hilarious

[Abhiraj @abhididthat]

@maxhodak_ this is awesome but do you think BioE is the most skill rewarding engineering degree for students going into med devices?? (coming from one myself)

[Abhiraj @abhididthat]

Just saw a homeless woman ask a homeless man to buy her a sandwich at McDonalds but didn’t ask a single average folk. Crazy social phenomenon that the homeless woman believed that the homeless man had the most humility out of all as only he could really understand her struggle

[Abhiraj @abhididthat]

Working alongside the PRIMA team is motivating as it gets. Excited to see PRIMA’s worldwide impact 🥳

[Science Corporation @ScienceCorp_]

Read more: https://t.co/1086EkkH6r

[Abhiraj @abhididthat]

@davieball @genomecomputer 🧬

[gabriel @gabriel1]

this took so long for me to understand: the bottleneck to more innovation is not more high intelligence people, but more people having an interest in hard problems

it's impossible to create new useful things if you don't get immense happiness from making that thing

[Abhiraj @abhididthat]

A lot of young engineers don’t realize working in deep tech is one of the best ways to learn. Dealing with the uncertainty of new technologies and the certainty of regulations you face pushes problem-solving like no other

[Abhiraj @abhididthat]

BCI may be one of the first technologies which can redefine what human-technology interaction and human health mean at the same time

[Yan Liberman @YanLiberman]

Most investors still frame neurotech around the ~$65B 2035 clinical TAM.

That market matters, but I think it understates the prize. The bigger question is whether BCIs become the next major human-machine interface layer: keyboards → touchscreens → thought.

The reason this feels investable now isn’t that the science suddenly appeared. It’s that the engineering stack is finally catching up: EEG hardware costs have collapsed, neural decoding is moving to low-power edge/implantable chips, and materials like PEDOT + graphene are already making their way into humans.

Neurotech has spent decades looking like “too early” frontier science. Increasingly, it looks like an underwritable platform shift.

Full Delphi thesis from @august_wstein and @DrewAHenderson:
https://t.co/L9f1YZupqO

[Abhiraj @abhididthat]

@Rinorg1 @CoachHoffSMCC Missing out if you don’t take a look‼️