sacha @SachaIchbiah Twitter profile

Pinned Tweet

sacha

@SachaIchbiah

2 months

time to lock in

0

1

7

Last Seen Profiles

@cukienaknikmati

@KairaKelli48164

@AspenReligion

@stwmaniax

@bokeplokalmalam

@LLhudiee

@LithsareKGu7T8

@swati_biology

@BlackShoppe2023

@BandarStw

@Isabella_Sexhot

@CaptainLEVI_XXX

@YuliaYulia131

@ngentot_stw

@pitiable_piece_

@Nagasakei

@jandakembangstw

@bokeplokalmalam

@heuilho

@Rivera9531

@lrshears

@tansdesty

@s1u1di13

@BinorRaja

@tansdesty

@Doll511Doge

@STR1kar17x

@dokge14

@bokeplokalmalam

@nilayts61

@CTravaris60299

@Isabella_Sexhot

@BinorRaja

@shani97732547

@bokeplokalmalam

sacha

@SachaIchbiah

11 months

Measuring 3D forces in early embryos at scale is now possible! Thrilled to share that my PhD paper has been published in @naturemethods . Huge congrats and thanks to my coauthors @RDumollard , @Alex_Mc_Dougall , @HerveTurlier for their incredible contributions!

Turlier lab

@virtual_embryo

11 months

Our approach to infer forces from 3D fluorescence microscopy is published in Nature Methods ! Congrats to all authors @SachaIchbiah @RDumollard @Alex_Mc_Dougall @HerveTurlier Simply type pip install foambryo to try And stay tuned... @CirbCdf @CNRSbiologie

6

67

261

3

28

183

sacha

@SachaIchbiah

1 year

Wow, we didn't expect so many reactions! As a result, @virtual_embryo and I have decided to start developing a @napari_imaging plugin to provide meshing and force inference for everyone! (Here, with C.elegans data from )

Turlier lab

@virtual_embryo

2 years

Can we predict cellular forces just from microscopy ? We are happy to release a robust method by @SachaIchbiah to create 3D atlases of relative cell mechanics in embryos or tissues from fluorescent images of cell outlines: A thread (1/n)

11

140

506

4

26

151

sacha

@SachaIchbiah

2 years

Very happy to introduce this big chunk of my PhD work 🥳

Turlier lab

@virtual_embryo

2 years

Can we predict cellular forces just from microscopy ? We are happy to release a robust method by @SachaIchbiah to create 3D atlases of relative cell mechanics in embryos or tissues from fluorescent images of cell outlines: A thread (1/n)

11

140

506

6

17

133

sacha

@SachaIchbiah

10 months

During the holiday, I played with two librairies that I love: - Outlines from the team of @dottxtai @remilouf , to do regex-guided text generation using LLMs - MLX, the ML framework optimized for Apple Silicon @awnihannun The result: Guided generation on LLMs running locally on a

2

8

69

sacha

@SachaIchbiah

10 months

@julesterpak

0

1

66

sacha

@SachaIchbiah

9 months

I never realized that Replit was being built in Replit. Like programming languages that are compiled with a compiler written in their own language, there is a beauty to that. Bootstrapping in action.

Amjad Masad

@amasad

9 months

Replit engineer @PhilMacEachron building Replit in Replit on the moon (in Vision Pro)

22

11

316

2

1

30

sacha

@SachaIchbiah

1 year

English is definitely the new hot programming language in image analysis🔥

Loïc A. Royer 💻🔬⚗️

@loicaroyer

1 year

🚨 #ChatGPT + @napari_imaging 🚨 Releasing my latest weekend project: Omega, an autonomous LLM agent that writes image processing and analysis code, fixes its mistakes, accesses the napari viewer, makes widgets, & more! @LangChainAI @OpenAI #OmegaAgent

17

259

905

0

3

22

sacha

@SachaIchbiah

9 months

New Outlines-MLX release ! pip install -U outlinesmlx outlinesmlx is a minimalistic library aimed at adapting the Outlines library for use with the MLX framework.. Both Outlines and MLX are evolving really fast. MLX models are now imported from HuggingFace with mlx-lm.

0

3

22

sacha

@SachaIchbiah

1 year

Statistical physicists would say that a great way to spot out-of-equilibrium processes is to see if the movie can be reversed in time and still be « plausible ». @martweig lab shows that this physical principle can be used on microscopy images to spot cell divisions 🤩

Martin Weigert

@martweig

1 year

Happy to share our latest paper on self-supervised representation learning for live-cell microscopy with time arrow prediction (TAP). Great work by @bengallusser and @max_stieber !

4

62

215

1

2

18

sacha

@SachaIchbiah

5 months

. @victor_plce , @louisdebenoist , and I are thrilled to announce our pre-seed round, led by @VentureF_VC with participation from K5 Global, @kimaventures , stemai( @ngranati ), @join_ef and @transposevc , alongside Olivier Pomel (Datadog), Florian Douetteau (Dataiku),

2

3

21

sacha

@SachaIchbiah

1 year

Thrilled to share our new publication in @CellCellPress where we delve into the modalities of membrane tension propagation in cells. A testament to the incredible collaboration with @OWeinerLab @BustamanteLab @virtual_embryo !

Henry De Belly

@Henry_de_belly

1 year

I am beyond excited that our story where we address the modalities of membrane tension propagation in cells is out @CellCellPress ! This is the fruit of an amazing collaboration with @OWeinerLab @BustamanteLab @virtual_embryo

12

53

230

1

16

sacha

@SachaIchbiah

2 years

Looks like vision is having his unsupervised-learning moment. The paper explains all the little tricks needed to make the magic work, very inspiring !

AI at Meta

@AIatMeta

2 years

Announced by Mark Zuckerberg this morning — today we're releasing DINOv2, the first method for training computer vision models that uses self-supervised learning to achieve results matching or exceeding industry standards. More on this new work ➡️

91

894

4K

1

0

14

sacha

@SachaIchbiah

6 months

Great to see @louisdebenoist pitch @getcubeai at our demo day @join_ef !

2

13

sacha

@SachaIchbiah

5 months

Wait FastAPI was created by just one person? How insanely cracked is that @tiangolo ?

1

0

12

sacha

@SachaIchbiah

5 months

@yacineMTB

0

11

sacha

@SachaIchbiah

5 months

The blog post ( @karpathy ) that inspired a thousand PhDs (including mine):

0

1

9

sacha

@SachaIchbiah

6 months

Writing your PhD Thesis VS Building your startup

1

0

9

sacha

@SachaIchbiah

6 months

@paulg That’s why tailwind rocks

0

8

sacha

@SachaIchbiah

10 months

In biomedical image analysis, the most effective algorithms still depend on traditional convolutional networks, such as U-Net, as long-range correlations makes Transformers inefficient. It’s exciting to see new architectures like Mamba 🐍 being applied in this field!

Bo Wang

@BoWang87

10 months

🎉 Thrilled to announce the very first 2024 preprint from our lab: U-Mamba 🐍, a novel backbone for universal bio-medical image segmentation! Background: Transformers have not transformed bio-medical images. While Transformers excel in general vision and language tasks, they

17

104

648

1

0

9

sacha

@SachaIchbiah

5 months

- had a call with a @Replit sales rep - the guy used the word « bloated » 3 times to describe competitors -Ok I’m in

0

10

sacha

@SachaIchbiah

1 year

@fchollet We all use PyTorch lol

0

8

sacha

@SachaIchbiah

5 months

@naderi_yeganeh Bro you just run (picture)

0

7

sacha

@SachaIchbiah

10 months

As @karpathy said to @lexfridman : In AI, we are leaving the era of benchtop science and entering the era of Large Hadron Collider science. A few centuries ago, physicists could conduct small-scale experiments in their labs and make groundbreaking discoveries. Now, it is becoming

hardmaru

@hardmaru

10 months

Google’s Gemini paper has ~1000 authors

44

142

1K

1

0

7

sacha

@SachaIchbiah

5 months

@yacineMTB I am a Vercelcel

0

6

sacha

@SachaIchbiah

11 months

For French folks: My PhD research has been featured in a vulgarisation article highlighted by the CNRS! Enjoy the read!

CIRB_CdF

@CirbCdf

11 months

| #research | #publication | 🗞️📣Cartographier la mécanique d'un embryon d'un seul coup de microscope @virtual_embryo @cirb @cnrs @cdf1530 pour en savoir plus : ▶️

0

4

11

0

6

sacha

@SachaIchbiah

5 months

Build, take feedback, iterate—do people realize this is just gradient descent applied to real life? This classic advice from @paulg / @ycombinator seems obvious, yet so many people fail to do the obvious things. Math describes the fabric of the universe. Its wisdom is

2

6

sacha

@SachaIchbiah

2 years

Awesome ! Things go so fast these days 🤯

Jordão Bragantini

@jobragantini

2 years

We created a fully-featured @napari_imaging plugin for @MetaAI 's incredible Segment Anything Model (SAM). It supports automatic instance segmentation and human feedback from points and bounding boxes. Try it! #segmentanything

6

77

375

0

6

sacha

@SachaIchbiah

1 year

Good moment to thank @RDumollard , for his patience and willingness to push the project forward on ascidians embryos. His feedback as a biologist was crucial for us to understand what insights people could learn from force-inference.

2

4

6

sacha

@SachaIchbiah

1 year

Don’t forget to delve into the details of our optimal transport-based flow inference algorithm in the supplementary note and on:

GitHub - sacha-ichbiah/optiflow: Reconstruction of flows from kymographs with optimal transport

Reconstruction of flows from kymographs with optimal transport - sacha-ichbiah/optiflow

github.com

1

6

sacha

@SachaIchbiah

6 months

Who missed this gem ?

KTO: Model Alignment as Prospect Theoretic Optimization

Kahneman & Tversky's $\textit{prospect theory}$ tells us that humans perceive random variables in a biased but well-defined manner (1992); for example, humans are famously loss-averse. We show...

arxiv.org

0

2

6

sacha

@SachaIchbiah

2 years

Very happy to have been part of this collaboration with @OWeinerLab and @BustamanteLab . I worked on new methods to infer flows occurring on cell membranes by solving an optimisation problem with optimal transport.

Henry De Belly

@Henry_de_belly

2 years

Really hyped to share this preprint of some of my postdoctoral work with @OWeinerLab @virtual_embryo @BustamanteLab in which we set out to investigate whether membrane tension propagation in cells Here is a summary of our findings 🧵

13

47

178

1

3

5

sacha

@SachaIchbiah

1 year

@zoink The great alignement

0

1

sacha

@SachaIchbiah

2 years

Great story ! Telescopes’ designs are always very clever and it is nice to see that microscopy can borrow ideas from this field 👌

Fabian F. Voigt

@voigtvision

2 years

From macro- to microcosm: A feature on the 🔭telescope & 🐚scallop- inspired #schmidtobjective #Microscope 🔬objective in @ScienceMagazine @NewsfromScience by @Sara_Reardon ! #Microscopy

4

24

97

0

5

sacha

@SachaIchbiah

5 months

My kids will have a synthesizer and will learn the basics of Fourier analysis and signal processing from that without even realising it

kache

@yacineMTB

5 months

hearing reports from dads about kids who have 3d printers, that they are learning CAD & out engineering their own fathers. i love the future so much

26

15

472

1

0

4

sacha

@SachaIchbiah

4 months

Morning read from @dnbt777 Life is a video game and you’d better be good at it

How to Build Anything Extremely Quickly - Learn How To Learn

Do "outline speedrunning": Recursively outline an MVP, speedrun filling it in, and only then go back and perfect. This is a ~10x speed up over the 'loading-bar' style (more on that below) Don't just...

learnhowtolearn.org

1

0

5

sacha

@SachaIchbiah

9 months

When I open a new window instead of a new tab it means things are getting really serious

0

5

sacha

@SachaIchbiah

6 months

@burny_tech We got this guy

1

0

5

sacha

@SachaIchbiah

6 months

@yacineMTB George Hotz ftw

0

1

5

sacha

@SachaIchbiah

1 year

@rohanpaul_ai @DrJimFan Nougat is from the same first author

0

5

sacha

@SachaIchbiah

2 years

Good initiative ! It is not all about money however, geographical instability is a big problem imo (compared to industry where you can work wherever your want)

Leslie Vosshall PhD

@leslievosshall

2 years

🧵📢 We need to retain talented postdocs in academic science. I'm happy to say that @HHMINEWS has taken a big step forward this week. $70,000 That's the new starting minimum salary for all postdocs employed in HHMI labs (1/4)

26

256

1K

1

0

5

sacha

@SachaIchbiah

6 months

I am starting to think that we will see the return of universal scientists—people who can leverage AI to quickly and deeply understand various topics and solve all sorts of problems. AI frees us from the burden of overspecialization. Now, we can go both deep AND broad.

0

5

sacha

@SachaIchbiah

6 months

@yacineMTB 100% based

0

4

sacha

@SachaIchbiah

1 year

@francoisfleuret Alphago is not clever it only predicts the next move ! Oh wait…

0

4

sacha

@SachaIchbiah

1 year

@ItaiYanai Wtf it’s more four or five per day 😂

0

4

sacha

@SachaIchbiah

10 months

In Paris, the density of talent, outstanding universities, research labs, and company headquarters is unparalleled in Europe. 🔥🔥 (Paris is so smaaaall!) When building there, one can witness a magical network effect: multiplying unexpected opportunities and serendipitous

Omar Sanseviero

@osanseviero

10 months

The ML ecosystem in France is on fire🔥 It has amazing talent and resources. Here are 10 facts you might not know: 1. There are great research labs - from @MistralAI and @kyutai_labs to large ones from @AIatMeta and @GoogleDeepMind . The Llama 2 and CodeLlama authors are based in

47

310

2K

0

4

sacha

@SachaIchbiah

11 months

Awesome picture 🙀

France-BioImaging Research Infrastructure

@Fr_BioImaging

11 months

🥈2nd Place: "Tardigrade embryos protected by mother's molt", Gonzalo QUIROGA-ARTIGAS, CRBM 🔬This image represents a tardigrade molt covering three developing embryos with confocal microscopy.

1

5

21

0

4

sacha

@SachaIchbiah

10 months

Google @GoogleDeepMind is somewhat lagging in the LLM distribution space, but it has other tricks up its sleeve 🧬

Nathan Benaich

@nathanbenaich

10 months

🧬 AI-first bio season looking tasty already! $82.5M upfront + up to $3B (and change) in biobucks for @IsomorphicLabs + @EliLillyandCo and @Novartis

6

14

195

0

4

sacha

@SachaIchbiah

6 months

Every day I thank God for inspiring humans to build such wonders

1

0

4

sacha

@SachaIchbiah

6 months

Zuck it is all in your hands now

Tsarathustra

@tsarnick

6 months

Sam Altman says the 10 most powerful AI systems in the world should have the equivalent of weapons inspectors

100

66

370

0

2

sacha

@SachaIchbiah

5 months

Embrace the flow state

Oren John

@orenmeetsworld

5 months

the creative way

58

2K

21K

0

2

4

sacha

@SachaIchbiah

2 years

@InternetH0F @DudespostingWs

0

4

sacha

@SachaIchbiah

5 months

Perhaps the role of researchers is to test new ideas on a small scale in a capital-efficient manner, and let large research labs scale them up if they prove promising

Irina Rish

@irinarish

5 months

It's mind-boggling that many AI researchers are still not quite aware of this Bitter Lesson. Unbelievably hard to overcome this mental attitude of sticking to small models/data and focusing on overly complex models/methods. Especially in academia...

31

40

291

1

0

4

sacha

@SachaIchbiah

9 months

Glad to see my @OutlinesOSS mlx adapter in the list of cool projects built with MLX ! pip install -U outlinesmlx

Awni Hannun

@awnihannun

9 months

Collected some of the amazing projects people are building with MLX in one place: Looking at that list, it's hard believe MLX is just 2 months old.

8

70

376

0

4

sacha

@SachaIchbiah

2 years

Very nice to see that we can still make progress on this 200,000-year-old grand challenge 😂

Cameron A. Schmidt, PhD

@CAS_ReproLab

2 years

Digging this @ScienceMagazine cover🤙🔥 Excited to dig in to the special section on human reproduction (p148)🤓 #ReproRocks

1

0

18

0

4

sacha

@SachaIchbiah

1 year

@Austen This hammer swing is that of someone who has never done manual work in his entire life

0

3

sacha

@SachaIchbiah

6 months

Plus, you come with all the computer engineering, linear algebra, group theory, functional analysis, and probability knowledge that is fundamental to building intuition about deep learning.

0

2

sacha

@SachaIchbiah

4 months

Science is a method, not a cult

anpaure

@anpaure

4 months

10

34

513

0

3

sacha

@SachaIchbiah

5 months

This is your daily reminder not to be an NPC

NIK

@ns123abc

5 months

Mr. Jobs dropped a massive redpill

39

925

5K

0

2

sacha

@SachaIchbiah

1 year

@fchollet I love reading your tweets, but I disagree with you on this one. The word emergence has been borrowed from statistical mechanics, where everybody use it all the time (see the classical P.W.Anderson article « More is different ») . Google Palm illustrates this concept quite well

1

0

3

sacha

@SachaIchbiah

6 months

@yacineMTB Woaw indeed. Scale invariance magic embedded in the weights themselves, fascinating

1

0

2

sacha

@SachaIchbiah

5 months

The voice interface is a distraction. GPT4o is really good a coding That’s the only thing that matters

0

3

sacha

@SachaIchbiah

8 months

When the model outputs a Nan in the middle of the training

0

3

sacha

@SachaIchbiah

9 months

Hidden gem on how to tackle data-science competitions from the chief data scientist of @OwkinScience (in French🇫🇷)

L'apprentissage face à la malédiction de la grande dimension (8) -...

Enseignement 2017-2018 : L'apprentissage face à la malédiction de la grande dimensionSéminaire du mercredi 21 février 2018 : S'attaquer à une compétition de ...

www.youtube.com

0

2

3

sacha

@SachaIchbiah

6 months

I’m always impressed by how technical the CEOs are in the Bay. In France, most of them are managers who know nothing about the tech

Yam Peleg

@Yampeleg

6 months

Zuck knows tons of technical details. I won't be surprised it he is working with the team by himself

18

10

494

0

3

sacha

@SachaIchbiah

5 months

Balaji is right. We love Huberman but we would love him even more if he was good at math. Don’t trust the experts. Don’t trust Huberman. Only trust yourself

Balaji

@balajis

6 months

HUBERMAN > HARVARD MAN François, I disagree and here’s why. ON THE SPECIFICS First, on the specifics — I don’t think one can “correct” someone without posting the correct answer. And the equation for the geometric CDF isn’t some random mumbo jumbo. It’s exactly what you need to

98

165

1K

0

3

sacha

@SachaIchbiah

2 years

Latent Space arithmetics tells is a lot about how these things work, mathematically speaking.

Jim Fan

@DrJimFan

2 years

One really cool thing you can do with diffusion models is latent space arithmetic. Remember the famous word2vec example? King - Man + Woman = Queen. You can do the same for image generation! Also works for (baby yoda) - baby + (old man) Demo by @genmoai @ajayj_ .

8

78

354

0

3

sacha

@SachaIchbiah

8 months

AIs can code, and this is great. The thing is, the most accurate way to explain to an AI what you want it to code is… through code itself. This idea was expressed by @stephen_wolfram as early as in 2019: « Why not just use natural language for everything? Just like in the

0

3

sacha

@SachaIchbiah

9 months

I love listening to music in the morning to remind myself that every artist once had the same high-risk, high-reward attitude I have right now. They all have been grinding hard in the shadows for a long time before thriving. Feels energizing.

0

3

sacha

@SachaIchbiah

6 months

@yacineMTB 😂😂

0

3

sacha

@SachaIchbiah

5 months

People are announcing AGI in 2027 but finishing their PhDs in 2028

0

3

sacha

@SachaIchbiah

2 years

Positivity rocks ! There are indeed a lot of nice things in a scientist life 😎

Burton Lab

@burtonlab

2 years

It's nice to see someone highlight all the positive aspects of being an academic scientist!

1

6

30

0

3

sacha

@SachaIchbiah

6 months

Pure legend

terminally onλine εngineer 🇺🇦

@tekbog

6 months

imagine showing up with eight of your best lads and just getting shit done

33

75

2K

0

3

sacha

@SachaIchbiah

6 months

Very interesting perspective by @ByrneHobart

Stripe and Solid-State Economics - by Byrne Hobart

Plus! Hulu's Regulatory Arbitrage; Post-Archegos; Patents; Emissions: a Global Problem

www.thediff.co

0

2

sacha

@SachaIchbiah

5 months

This is so cool. We can feel those old-school internet vibes again

mel

@vmfunc

5 months

LADIES AND GENTLEMAN WE GOT IT BACK

43

97

1K

0

2

sacha

@SachaIchbiah

9 months

Looks like this team is cooking something great ! 🔥

Rémi 📎

@remilouf

9 months

Please retweet, the worlds needs to know 😊

1

6

42

1

0

2

sacha

@SachaIchbiah

4 months

LLM + biology is a really interesting field Some people are cooking 🫢

0

2

sacha

@SachaIchbiah

6 months

If beauty is subjective, why do all dashboards look like Stripe’s ?

0

2

sacha

@SachaIchbiah

6 months

Jim Simons' philanthropic organization supports some of the most amazing research in physics and basic science. He was a unique genius, and a true legend. RIP Jim. 🫡

Matt

@mprkhrst

6 months

> independently discover a Zeno's paradox at age 3 > MIT at 17, grad level math in 1st year > graduate in 3 years > drive motor scooters from Boston to Bogotá with the boys > start a company in Colombia > start code breaking with the IDA for money > solve minimal varieties in

217

3K

26K

0

2

sacha

@SachaIchbiah

5 months

Technology has given us everything we have

YiMing

@yimingdothan

5 months

3

27

269

0

1

sacha

@SachaIchbiah

9 months

There is no negative feedback on LinkedIn. You cannot discuss about science there 😅

Bo Wang

@BoWang87

9 months

I’m genuinely curious about the most popular social media platform among scientists. My personal experience suggests LinkedIn is overtaking Twitter. I’d love to hear broader opinions on this 🙏

7

2

12

0

2

sacha

@SachaIchbiah

7 months

I need this for my company 🙀

Patrick Blumenthal

@PatrickJBlum

7 months

Anduril Japan Recruiting Ad, 2025.

30

101

1K

0

1

sacha

@SachaIchbiah

7 months

Morning read from @packyM

Techno-Optimistic Media

A Not Boring x First Principles Guide to Modern Startup Storytelling

www.notboring.co

0

1

sacha

@SachaIchbiah

6 months

@FlohOfWoe Not all heroes wear capes

0

2

sacha

@SachaIchbiah

2 years

@MorphoTweet @EntraCod @sentinal4d You can ask for the source code but please don’t ask why our model organism is called Phallusia

1

0

2

sacha

@SachaIchbiah

2 years

@vectorgen What is the specific reason ?

0

2

sacha

@SachaIchbiah

6 months

You may need to be trained in math and physics to think from first principles.

1

0

2

sacha

@SachaIchbiah

6 months

@yacineMTB Have you heard of @keenanisalive ? He is doing really cool stuff

0

2

sacha

@SachaIchbiah

1 year

Exciting times ahead !

clem 🤗

@ClementDelangue

1 year

I believe we need open-source alternatives to ChatGPT for more transparency, inclusivity, accountability and distribution of power. Excited to introduce HuggingChat, an open-source early prototype interface, powered by OpenAssistant, a model that was released a few weeks ago.

64

410

2K

0

1

sacha

@SachaIchbiah

1 year

@haesleinhuepf @virtual_embryo @napari_imaging Not yet aha this is still work in progress 🙈 We will release it when the paper is published

1

0

2

sacha

@SachaIchbiah

6 months

@yacineMTB Don’t worry Zuck is gonna win

0

2

sacha

@SachaIchbiah

10 months

The MLX team ships daily ! Their framework is very simple to use (for a proficient PyTorch user) and FAST 😻

Awni Hannun

@awnihannun

10 months

Some great notes on fine tuning Mistral 7B with MLX: “MLX is fast. This took 3:24 hours as opposed to almost 12 hours using llama.cpp” And its going to keep getting faster 🚀

2

20

285

0

2

sacha

@SachaIchbiah

6 months

An unreasonable fraction of the most driven people I know get all their motivation from this guy

1

0

2

sacha

@SachaIchbiah

9 months

When I first learned CUDA (to implement a custom kernel for performing Fourier transforms on meshes), the documentation was very poor and debugging was a mess. Still, it was a very thrilling feeling to unleash this compute power. Now, with AI-assisted coding and all the

Jeremy Howard

@jeremyphoward

9 months

I used to find writing CUDA code rather terrifying. But then I discovered a couple of tricks that actually make it quite accessible. In this video I introduce CUDA in a way that will be accessible to Python folks, & I even show how to do it all in Colab!

21

365

3K

0

2

sacha

@SachaIchbiah

2 years

@G_Charras @Kazu_YAMAMOTO4 @virtual_embryo And as a bonus, we made the force-inference pipeline freely available on GitHub: Note that the methods employed generalized very well in 3D sooo.. stay tuned !

GitHub - VirtualEmbryo/foambryo2D: Python package to infer tensions from 2D microscopy images of...

Python package to infer tensions from 2D microscopy images of embryos and tissues - VirtualEmbryo/foambryo2D

github.com

0

1

2

sacha

@SachaIchbiah

2 years

Very happy to see this out ! Thanks to @Kazu_YAMAMOTO4 @G_Charras @HerveTurlier and all the others involved for giving me the opportunity to do this work !

Guillaume Charras

@G_Charras

2 years

What forces underlie the cell arrangement of early C. elegans? @Kazu_YAMAMOTO4 & @SachaIchbiah dissected the forces shaping C. elegans embryos in a great collaboration between @G_Charras & @virtual_embryo labs: A brief summary (1/4).

2

10

62

1

0

2

sacha

@SachaIchbiah

6 months

@BlackHC @queirozfcom Berserk indeed

1

0

2

sacha

@SachaIchbiah

1 year

@aml78i Thanks ! On which data do you plan to do force inference ? We can talk in dm

1

0

2

sacha

@SachaIchbiah

7 months

@TrentonBricken and @_sholtodouglas explain that the ability to execute without letting themselves get blocked is what sets them appart. This is something I have witnessed during my PhD. A lot of cool ideas are floating around and this is often your ability to execute rapidly to

0

2

sacha

@SachaIchbiah

8 months

Trying to explain to my co-founders why I think Nvidia isn't worth more than Google or Amazon. My two cents as a CUDA, PyTorch, and ML developer: - Nvidia is a great company (let's start with that). - I will probably buy some A100s later this year. - CUDA, their C++-like

0

2