Bruno Gavranović✓ Add us as a preferred source on Google
🔄 Data last refreshed 1 month ago
🤝
Make a deal with Bruno Gavranović
🔒 Paid proposals held safely in escrow — released only when the work's approved.
Followers
405
Account age
3 yrs
💰 Suggested sponsorship rate
$40 – $65 / sponsored post
Estimated from audience size & typical reach — every deal is negotiable and held in SocialDB escrow.
🎁 Free analysisfor Bruno Gavranović
Known for
17 views
I just published a new blog post!
Types and Neural Networks
(https://www.brunogavranovic.com/posts/2026-04-20-types-and-neural-networks.html)
It's about what comes into view once dependent type systems become a part of neural network training.
9 views
I never wrote about it here, but as of some time ago I figured out a basic implementation for named axes in TensorType:
https://github.com/bgavran/TensorType
This means that now you're:
a) forced to assign some meaning to all your axes
b) cannot by accident sum over sequence length, for instance, instead of "batch"
There's still a long way to go to get this fully integrated, but I'm quite excited
6 views
Almost forgot to mention:
I'll be presenting TensorType today at 16:30 EEST at the ACT conference
https://actconf2026.github.io/programme.html
1 views
This also solved a few technical issues behind the scenes, and eased ergonomics around distinguishing between creation of cubical tensors (which I want to have backward compatibility with), and non-cubical tensors (which this framework enables).
That is, compare the syntax for creating a non-cubical tensor (image attached to this post) vs one for non-cubical ones (image in the original post)
📊 Post engagement
Recent posts · likes + reposts
Apr 14latest
7
Avg engagement / post
2–9
Typical post (engagement)
1.7%
Engagement vs followers
63/100
Consistent reach
Dec 2022
On Mastodon since
Half of recent posts land between 2 and 9 likes + reposts (median 4) — a dependable floor a sponsored post can count on.
🔥 Top post: I just published a new blog post! Types and Neural Networks (htt · 17 likes + reposts
📊 Activity & format
Posting cadence
0.57 / week
A lower-frequency account — each post lands with more weight.
Content mix
Mostly text
Recent: 8 text · 4 image · 0 video.
Follower / following
4.3×
Follows 94 back. A more reciprocal / networked account.
Recent posts
View on Mastodon ↗
🔥 Top post
I just published a new blog post!
Types and Neural Networks
(https://www.brunogavranovic.com/posts/2026-04-20-types-and-neural-networks.html)
It's about what comes into view once dependent type systems become a part of neural network training.
RE: https://mathstodon.xyz/@bgavran/116660321337179277
Regarding recent news
Finally added a reinforcement learning section to the list: https://github.com/bgavran/Category_Theory_Machine_Learning
I'm sure I'm missing a few papers
Almost forgot to mention:
I'll be presenting TensorType today at 16:30 EEST at the ACT conference
https://actconf2026.github.io/programme.html
RE: https://mathstodon.xyz/@julesh/116658899165675527
C3, but with a caveat:
The negative externalities the row A highlights are real, and not to be dismissed. However, these are not a fundamental aspect of this technology, but at the mome…
For me, this answered a long-standing question about works like CHAD (https://arxiv.org/abs/2103.15776) or The Differentiable Curry (https://dimitriv.github.io/papers/hoad-workshop.pdf) which allow us to differentiate through structured ob…
This has been on my mind for a while, and its something we've been getting quite excited about at GLAIVE.
It started as narrow question:
> How can we train a network to generate a dependent pair in a way that is *correct-by-construction*?
…
New blog post coming up soon!
I'm very excited about some of the work we've been doing at GLAIVE about making the output space of frontier models *typed*.
Stay tuned :)
What is really exciting is that *type-safe einsum* is now fully within reach, and I have a pretty good idea how to implement it.
On the technical side it finally enabled tensor reshapes to fully fall out of the categorical machinery: they're extensions of morphisms of containers.
That is, tensor reshape defined here: https://github.com/bgavran/TensorType/blob/main/s…
This also solved a few technical issues behind the scenes, and eased ergonomics around distinguishing between creation of cubical tensors (which I want to have backward compatibility with), and non-cubical tensors (which this framework ena…
I never wrote about it here, but as of some time ago I figured out a basic implementation for named axes in TensorType:
https://github.com/bgavran/TensorType
This means that now you're:
a) forced to assign some meaning to all your axes
b) …
Almost forgot to mention:
I'll be presenting TensorType today at 16:30 EEST at the ACT conference
https://actconf2026.github.io/programme.html
RE: https://mathstodon.xyz/@julesh/116658899165675527
C3, but with a caveat:
The negative externalities the row A highlights are real, and not to be dismissed. However, these are not a fundamental aspect of this technology, but at the mome…
For me, this answered a long-standing question about works like CHAD (https://arxiv.org/abs/2103.15776) or The Differentiable Curry (https://dimitriv.github.io/papers/hoad-workshop.pdf) which allow us to differentiate through structured ob…
This has been on my mind for a while, and its something we've been getting quite excited about at GLAIVE.
It started as narrow question:
> How can we train a network to generate a dependent pair in a way that is *correct-by-construction*?
…
New blog post coming up soon!
I'm very excited about some of the work we've been doing at GLAIVE about making the output space of frontier models *typed*.
Stay tuned :)
What is really exciting is that *type-safe einsum* is now fully within reach, and I have a pretty good idea how to implement it.
On the technical side it finally enabled tensor reshapes to fully fall out of the categorical machinery: they're extensions of morphisms of containers.
That is, tensor reshape defined here: https://github.com/bgavran/TensorType/blob/main/s…
This also solved a few technical issues behind the scenes, and eased ergonomics around distinguishing between creation of cubical tensors (which I want to have backward compatibility with), and non-cubical tensors (which this framework ena…
I never wrote about it here, but as of some time ago I figured out a basic implementation for named axes in TensorType:
https://github.com/bgavran/TensorType
This means that now you're:
a) forced to assign some meaning to all your axes
b) …
🐘 Community & instance
Home server
mathstodon.xyz
Their home server on the fediverse — the instance a creator picks signals the community they belong to.
✅ Link-verified
Verified link
Proved ownership of a website linked on their profile — Mastodon's green-check verification, a real identity signal rather than a paid badge.
On Mastodon since
Dec 2022
Joined in the Twitter-exodus wave of late 2022 — part of the migration that made Mastodon a real destination.
💡 Facts
🗓️Joined Mastodon in 2022 — 3 years ago.
👁️Averages 7 views per post.
📤Posts about 0.6× per week.
🕵️ Fake follower check
Estimated75/100
Good
Credibility score
92%
Real
Real audience
Low
Fake-follower risk
High
Data confidence
- Est. 92% real, active audience · Low fake-follower risk.
- Engagement (~1.7% of followers engage each post) is around typical for Mastodon.
- Verified account.
- Established account (3+ years old).
Heuristic estimate from engagement, follower ratios, account age & growth — a screening signal, not a guarantee.
About
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
📸 Gallery
🔀 Audience overlap
EstimatedEstimated shared audience with similar creators — useful for avoiding overlap (or doubling down) when planning a campaign.
SpookyEmily 🏳️🌈🏳️⚧️🎃👻
18%
Irish Mythology Podcast
16%
Marquis de Geek
15%
Anna Wilson 🏳️⚧️
15%
The Penguins In The Wall
10%
Doctor Spork
10%
More like this
Find more →
Mastodon
＋
Marquis de Geek ✓
@marquisdegeek@ohai.social
405
Followers
Mastodon
＋
The Penguins In The Wall
@glasgowpenguins@mastodon.scot
405
Followers
Mastodon
＋
Doctor Spork
@doctorspork@dice.camp
405
Followers
Mastodon
＋
SpookyEmily 🏳️🌈🏳️⚧️🎃👻
@emilyenough@hachyderm.io
405
Followers
Mastodon
＋
Irish Mythology Podcast
@irishmythology@mastodon.ie
405
Followers
Mastodon
＋
Anna Wilson 🏳️⚧️
@yesitsanna@hachyderm.io
405
Followers
Frequently asked questions
How much does Bruno Gavranović charge for a sponsorship?
Bruno Gavranović hasn't published fixed prices yet. Send a proposal through SocialDB and agree terms directly — your payment is held in escrow until the work is approved.
How do I contact Bruno Gavranović for a brand deal or collaboration?
Send a paid deal or collab request right here on SocialDB — we notify Bruno Gavranović, and once they accept, your payment is held safely in escrow and released when the work is approved. You don't need to track down an email.
How many followers does Bruno Gavranović have?
Bruno Gavranović has 405 followers on Mastodon.
✉ Message Bruno Gavranović
Reaching out to influencers is a Pro feature. Upgrade to message any influencer directly — perfect for brands and agencies booking sponsorships.
- ✓ Message any influencer from their listing
- ✓ The influencer gets notified by email
- ✓ Manage every conversation in one inbox
Already Pro? Log in.
🎤 Event / appearance with Bruno Gavranović
Booking an event / appearance is a Pro feature. Upgrade to book Bruno Gavranović for an in-person or virtual appearance — payment held safely in escrow until the event is done.
- ✓ Book them for events, livestreams, panels & more
- ✓ Bruno Gavranović gets notified by email
- ✓ Fee held in escrow, released after the appearance
Already Pro? Log in.
🚀
You're out of free requests this month
Free accounts get 5 per month. Go Pro for unlimited sponsor pitches, collab requests & sponsorship deals — plus featured placement, the Verified badge, free withdrawals and more.
Upgrade to Pro — $9.95/mo →Your free limit resets on the 1st of next month.
Auto-generated from Mastodon's public data — no affiliation with or endorsement by SocialDB. Is this you? Claim it · Remove