Bruno Gavranović
@bgavran.bsky.social
I'm building neural networks that generate provably correct code, and the software infrastructure for training them.
Recently experimenting with TensorType: https://github.com/bgavran/TensorType
www.brunogavranovic.com
Recently experimenting with TensorType: https://github.com/bgavran/TensorType
www.brunogavranovic.com
Reposted by Bruno Gavranović
The transition from “AI can’t do novel science” to “of course AI does novel science” will be like every other similar AI transition.
First the over-enthusiastic claims that are debunked, then smart people use AI to help them, then AI starts to do more of the work, then minor discoveries, & then…
First the over-enthusiastic claims that are debunked, then smart people use AI to help them, then AI starts to do more of the work, then minor discoveries, & then…
February 14, 2026 at 2:21 PM
The transition from “AI can’t do novel science” to “of course AI does novel science” will be like every other similar AI transition.
First the over-enthusiastic claims that are debunked, then smart people use AI to help them, then AI starts to do more of the work, then minor discoveries, & then…
First the over-enthusiastic claims that are debunked, then smart people use AI to help them, then AI starts to do more of the work, then minor discoveries, & then…
Reposted by Bruno Gavranović
there is a widespread belief among people with even a little technical and scientific savvy that Alchemy was a bunch of hooey, that they never published their experiments, that alchemists had to each discover on their own not to drink mercury, etc
but this is urban legend!
but this is urban legend!
February 14, 2026 at 12:20 AM
there is a widespread belief among people with even a little technical and scientific savvy that Alchemy was a bunch of hooey, that they never published their experiments, that alchemists had to each discover on their own not to drink mercury, etc
but this is urban legend!
but this is urban legend!
Reposted by Bruno Gavranović
This is not what is happening at all.
The amount of misinformation on BlueSky about AI is insane, and it keeps promising that AI is all hype that is going away soon.
A really dangerous position that cedes all AI policy and decisions about how it will be used to others.
Also Futurism is clickbait
The amount of misinformation on BlueSky about AI is insane, and it keeps promising that AI is all hype that is going away soon.
A really dangerous position that cedes all AI policy and decisions about how it will be used to others.
Also Futurism is clickbait
February 6, 2026 at 9:15 PM
This is not what is happening at all.
The amount of misinformation on BlueSky about AI is insane, and it keeps promising that AI is all hype that is going away soon.
A really dangerous position that cedes all AI policy and decisions about how it will be used to others.
Also Futurism is clickbait
The amount of misinformation on BlueSky about AI is insane, and it keeps promising that AI is all hype that is going away soon.
A really dangerous position that cedes all AI policy and decisions about how it will be used to others.
Also Futurism is clickbait
I'm quite proud of how far I've been able to get with TensorType: https://github.com/bgavran/TensorType
What started out as a casual "I wonder if I can implement type-safe tensors" question has now evolved into a fully-fledged library
What started out as a casual "I wonder if I can implement type-safe tensors" question has now evolved into a fully-fledged library
GitHub - bgavran/TensorType: Framework for type-safe pure functional and non-cubical tensor processing, written in Idris 2
Framework for type-safe pure functional and non-cubical tensor processing, written in Idris 2 - bgavran/TensorType
github.com
February 2, 2026 at 8:34 PM
I'm quite proud of how far I've been able to get with TensorType: https://github.com/bgavran/TensorType
What started out as a casual "I wonder if I can implement type-safe tensors" question has now evolved into a fully-fledged library
What started out as a casual "I wonder if I can implement type-safe tensors" question has now evolved into a fully-fledged library
If you're interested in some of the design choices behind TensorType, have a look at the great blog post that @Andrev just posted:
types.pl/@Andrev/1159...
TLDR; Tensors in NumPy are secretly built out of the composition product of containers
types.pl/@Andrev/1159...
TLDR; Tensors in NumPy are secretly built out of the composition product of containers
Andre Videla (@[email protected])
Glaive has a new blog post aimed at curious engineers https://glaive-research.org/2026/01/21/Generalised-tensors.html
types.pl
January 22, 2026 at 9:41 PM
If you're interested in some of the design choices behind TensorType, have a look at the great blog post that @Andrev just posted:
types.pl/@Andrev/1159...
TLDR; Tensors in NumPy are secretly built out of the composition product of containers
types.pl/@Andrev/1159...
TLDR; Tensors in NumPy are secretly built out of the composition product of containers
Reposted by Bruno Gavranović
Dear USAans, ICE will not let you complete an electoral process which might result in a government that might hold them to account. If you want your democracy back, you have to get rid of them *first*. Abolish ICE, you say? How, I ask?
January 20, 2026 at 9:08 PM
Dear USAans, ICE will not let you complete an electoral process which might result in a government that might hold them to account. If you want your democracy back, you have to get rid of them *first*. Abolish ICE, you say? How, I ask?
Reposted by Bruno Gavranović
On the drive home I was idly thinking about what changes I'd make, if I could, to our system of governance after this administration is - ideally - gone.
I suppose in no particular order, here is a list of what I'd do, sorted by the mechanism for doing it.
I suppose in no particular order, here is a list of what I'd do, sorted by the mechanism for doing it.
January 14, 2026 at 2:13 AM
On the drive home I was idly thinking about what changes I'd make, if I could, to our system of governance after this administration is - ideally - gone.
I suppose in no particular order, here is a list of what I'd do, sorted by the mechanism for doing it.
I suppose in no particular order, here is a list of what I'd do, sorted by the mechanism for doing it.
If you've been curious what I've been up to, the recently published report from GLAIVE reveals a part of it:
https://glaive-research.org/2025/12/08/q4-report.html
https://glaive-research.org/2025/12/08/q4-report.html
Q4 2025 report
Glaive Research Report Q4 2025
glaive-research.org
December 8, 2025 at 8:29 PM
If you've been curious what I've been up to, the recently published report from GLAIVE reveals a part of it:
https://glaive-research.org/2025/12/08/q4-report.html
https://glaive-research.org/2025/12/08/q4-report.html
Different products of containers, with examples on the List container:
List ⊗ List -> rectangular array
List ∘ List -> ragged array
List × List -> two lists
List + List -> a boolean value and a list
List ⊗ List -> rectangular array
List ∘ List -> ragged array
List × List -> two lists
List + List -> a boolean value and a list
November 17, 2025 at 7:36 PM
Different products of containers, with examples on the List container:
List ⊗ List -> rectangular array
List ∘ List -> ragged array
List × List -> two lists
List + List -> a boolean value and a list
List ⊗ List -> rectangular array
List ∘ List -> ragged array
List × List -> two lists
List + List -> a boolean value and a list
Reposted by Bruno Gavranović
Remember kids, theoretical physicists have irreparably ruined public trust in science by saying they *hope* we *could* find BSM evidence at the LHC, but Elon gets to predict his products will "do the sci-fi" and be wrong 100 times per year for profit but he's just giving them hope
The brave renegade truth teller: "the people must be made to believe obvious lies to pacify them"
November 17, 2025 at 12:21 PM
Remember kids, theoretical physicists have irreparably ruined public trust in science by saying they *hope* we *could* find BSM evidence at the LHC, but Elon gets to predict his products will "do the sci-fi" and be wrong 100 times per year for profit but he's just giving them hope
Had an absolutely fantastic time in Scotland. It's been great seeing a lot of friendly faces, and making exciting research progress with Glaive
November 15, 2025 at 10:36 AM
Had an absolutely fantastic time in Scotland. It's been great seeing a lot of friendly faces, and making exciting research progress with Glaive
London ---🚅---> Glasgow
November 10, 2025 at 11:59 AM
London ---🚅---> Glasgow
Containers and quotient containers can be interpreted as choices of a category S with some constraints on it, and a functor S -> Set.
For an ordinary container, S has to be a discrete category, i.e. a set.
For an ordinary container, S has to be a discrete category, i.e. a set.
November 2, 2025 at 10:40 PM
Containers and quotient containers can be interpreted as choices of a category S with some constraints on it, and a functor S -> Set.
For an ordinary container, S has to be a discrete category, i.e. a set.
For an ordinary container, S has to be a discrete category, i.e. a set.
October 31, 2025 at 11:28 AM
Stumbled upon this on HN:
"Categorical foundations for CuTe layouts"
https://research.colfax-intl.com/categorical-foundations-for-cute-layouts/
"Categorical foundations for CuTe layouts"
https://research.colfax-intl.com/categorical-foundations-for-cute-layouts/
Categorical Foundations for CuTe Layouts
In GPU programming, performance depends critically on how data is stored and accessed in memory. While the data we care about is typically multi-dimensional, the GPU’s memory is fundamentally one-d…
research.colfax-intl.com
September 22, 2025 at 8:49 PM
Stumbled upon this on HN:
"Categorical foundations for CuTe layouts"
https://research.colfax-intl.com/categorical-foundations-for-cute-layouts/
"Categorical foundations for CuTe layouts"
https://research.colfax-intl.com/categorical-foundations-for-cute-layouts/
Stumbled upon the paper "Sorting with Bialgebras and Distributive Laws" ( https://www.cs.ox.ac.uk/people/nicolas.wu/papers/Sorting.pdf )
which introdued me to the idea that certain sorting algorithms can be considered 'duals' of others:
which introdued me to the idea that certain sorting algorithms can be considered 'duals' of others:
September 18, 2025 at 2:22 PM
Stumbled upon the paper "Sorting with Bialgebras and Distributive Laws" ( https://www.cs.ox.ac.uk/people/nicolas.wu/papers/Sorting.pdf )
which introdued me to the idea that certain sorting algorithms can be considered 'duals' of others:
which introdued me to the idea that certain sorting algorithms can be considered 'duals' of others:
I have been thinking about fixpoints of containers recently, and noticed that there seems to be a close correspondence between fixpoints of extensions of containers \mu(Ext(C)) and iterated compositions I + C + C o C + C o C o C ...
Has this been formally established somewhere?
Has this been formally established somewhere?
September 16, 2025 at 11:34 AM
I have been thinking about fixpoints of containers recently, and noticed that there seems to be a close correspondence between fixpoints of extensions of containers \mu(Ext(C)) and iterated compositions I + C + C o C + C o C o C ...
Has this been formally established somewhere?
Has this been formally established somewhere?
Interestingly, this is still a question I do not have a good answer for, and a question that I didn't see tackled by the deep learning community at all.
September 8, 2025 at 6:47 PM
Interestingly, this is still a question I do not have a good answer for, and a question that I didn't see tackled by the deep learning community at all.
Reposted by Bruno Gavranović
Unbelievable!
Today, San Diego witnessed a government-sanctioned kidnapping. ICE stormed Buona Forchetta in military gear ripping workers away, then unleashing tear gas and smoke on the terrified community. Full on police state, fascism now used on American citizens. Stay aware.
#OnYourSide
Today, San Diego witnessed a government-sanctioned kidnapping. ICE stormed Buona Forchetta in military gear ripping workers away, then unleashing tear gas and smoke on the terrified community. Full on police state, fascism now used on American citizens. Stay aware.
#OnYourSide
May 31, 2025 at 4:04 AM
Unbelievable!
Today, San Diego witnessed a government-sanctioned kidnapping. ICE stormed Buona Forchetta in military gear ripping workers away, then unleashing tear gas and smoke on the terrified community. Full on police state, fascism now used on American citizens. Stay aware.
#OnYourSide
Today, San Diego witnessed a government-sanctioned kidnapping. ICE stormed Buona Forchetta in military gear ripping workers away, then unleashing tear gas and smoke on the terrified community. Full on police state, fascism now used on American citizens. Stay aware.
#OnYourSide
Stumbled upon a peculiar paper:
Aggregating time-series and image data:
functors and double functors
https://arxiv.org/abs/2504.05274
Aggregating time-series and image data:
functors and double functors
https://arxiv.org/abs/2504.05274
May 19, 2025 at 3:33 PM
Stumbled upon a peculiar paper:
Aggregating time-series and image data:
functors and double functors
https://arxiv.org/abs/2504.05274
Aggregating time-series and image data:
functors and double functors
https://arxiv.org/abs/2504.05274
I've always wanted someone to write a type-safe framework for tensor manipulation that
a) fully leverages first-class types, and
b) is ergonomic
Maybe that someone has to be me; I wrote a proof-of-concept in Idris and used it to implement transformers:
github.com/bgavran/Type...
a) fully leverages first-class types, and
b) is ergonomic
Maybe that someone has to be me; I wrote a proof-of-concept in Idris and used it to implement transformers:
github.com/bgavran/Type...
GitHub - bgavran/TypeSafe_Tensors: Framework for type-safe pure functional tensor processing, supporting non-cubical tensors
Framework for type-safe pure functional tensor processing, supporting non-cubical tensors - bgavran/TypeSafe_Tensors
github.com
May 7, 2025 at 11:07 AM
I've always wanted someone to write a type-safe framework for tensor manipulation that
a) fully leverages first-class types, and
b) is ergonomic
Maybe that someone has to be me; I wrote a proof-of-concept in Idris and used it to implement transformers:
github.com/bgavran/Type...
a) fully leverages first-class types, and
b) is ergonomic
Maybe that someone has to be me; I wrote a proof-of-concept in Idris and used it to implement transformers:
github.com/bgavran/Type...
Reposted by Bruno Gavranović
London --🚄--> Edinburgh
April 6, 2025 at 4:54 PM
London --🚄--> Edinburgh
Stumbled upon an Idris implementation of type-safe tensor computation (alongside many other things)
https://github.com/joelberkeley/spidr/tree/master
https://github.com/joelberkeley/spidr/tree/master
GitHub - joelberkeley/spidr: Accelerated machine learning with dependent types
Accelerated machine learning with dependent types. Contribute to joelberkeley/spidr development by creating an account on GitHub.
github.com
March 25, 2025 at 8:44 PM
Stumbled upon an Idris implementation of type-safe tensor computation (alongside many other things)
https://github.com/joelberkeley/spidr/tree/master
https://github.com/joelberkeley/spidr/tree/master