Sign in

Szymon Kaliski

@szymonkaliski.com
1.2K followers 331 following 150 posts

independent consultant, designing and developing computational interfaces; currently prototyping at Google Research and Google Creative Lab szymonkaliski.com

PostsRepliesMedia
Szymon Kaliski @szymonkaliski.com · 28/09/2026
✉️ Q3 2026 — Play with Putty, and a Quiet Summer szymonkaliski.com/newsletter/2...
szymonkaliski.com
Q3 2026
Play with Putty, and a Quiet Summer
000
Szymon Kaliski @szymonkaliski.com · 25/08/2026
excited to introduce one of the projects I helped create at Google -- labs.google/playwithputty -- I can't wait to share more in the coming weeks, for now you can sign up to the waitlist, we'll start bringing in people soon!
0191
Szymon Kaliski @szymonkaliski.com · 06/08/2026
what are the long term complications of being lean and muscular I wonder 🤔
120
Szymon Kaliski @szymonkaliski.com · 06/08/2026
also on the original quoted post, maybe not super outrageous but classics for me are "bmi says you are obese and should move more" and "are you on steroids" -- most fun if both land during the same visit
120
Szymon Kaliski @szymonkaliski.com · 06/08/2026
haha yes that's why I usually try to find "sport doctors" for these things
100
Szymon Kaliski @szymonkaliski.com · 06/08/2026
short of EMS is this even possible
120
Reposted by Szymon Kaliski
Rude1 Haunted Badness. ⁂ @rude1.blacksky.team · 27/07/2026
Hello. @mozilla.org is in the market for a "0→1 engineer" for a consulting role focused on open source AI 🌍 remote friendly 💰 budget ≈ $200K–$227,250 🕰️ fixed term contract (not FTE role) 🧑🏾‍💻 fluency in Python and/or TypeScript 🤖 comfortable working w/ agents 📧 contact mike@mozilla.org if interested
17227
Reposted by Szymon Kaliski
andrew blinn @disconcision.com · 11/07/2026
all will become fidgetable
1630437
Szymon Kaliski @szymonkaliski.com · 11/07/2026
this is fantastic Andrew
050
Szymon Kaliski @szymonkaliski.com · 29/06/2026
oh one more thing - I now often stage screenshots in a way that tries to provide more context in the image - leave a little note next to thing I'm screenshotting, a reference material, etc. - which makes it easier to find later / stumble upon randomly
020
Szymon Kaliski @szymonkaliski.com · 29/06/2026
it's one of those things like keeping a journal or zettlekasten or something - it's been more and more useful the more stuff is in it
110
Szymon Kaliski @szymonkaliski.com · 29/06/2026
I have two pretty common use cases - I either vaguely remember something ("Adam had this good bsky post about X") and want to get it back, or I want like ultra-curated Pinterest - I start with something fuzzily related to what I'm working on, and use similar images to dig out things I forgot about
120
Szymon Kaliski @szymonkaliski.com · 29/06/2026
love the question! I had a handful of situations where I wanted to come back to something and the page was gone or completely changed, removing the original text
100
Szymon Kaliski @szymonkaliski.com · 29/06/2026
✉️ Q2 2026 — Keeping Archivist Alive, and Other Side-Projects szymonkaliski.com/newsletter/2...
1120
Szymon Kaliski @szymonkaliski.com · 07/05/2026
happy birthday Sean!
140
Szymon Kaliski @szymonkaliski.com · 24/04/2026
time sure does fly
060
Szymon Kaliski @szymonkaliski.com · 18/04/2026
githubnext.com/projects/age...
githubnext.com
GitHub Next | Agentic Workflows
GitHub Next Project: Towards Natural‑Language Programming for GitHub Actions
120
Szymon Kaliski @szymonkaliski.com · 13/04/2026
fantastic talk!
060
Szymon Kaliski @szymonkaliski.com · 08/04/2026
maybe a bit too hard yea? I couldn't read anything serious while training, but I can take a note lol
100
Szymon Kaliski @szymonkaliski.com · 08/04/2026
mas.to/@szymon_k/11...
mas.to
Szymon Kaliski (@szymon_k@mas.to)
@jack@berlin.social @neauoire@merveilles.town my mental shortcut about this has been that there's a single nervous system in your body, which explains why my training sucks when my mind is elsewhere /...
100
Szymon Kaliski @szymonkaliski.com · 30/03/2026
✉️ Q1 2026 — Home Server on NixOS, Sandboxing in MicroVMs, and Feedback Loops for LLMs szymonkaliski.com/newsletter/2...
0180
Szymon Kaliski @szymonkaliski.com · 26/03/2026
cross site cookies should be bridges
010
Szymon Kaliski @szymonkaliski.com · 26/03/2026
blobs should be shaped and positioned on the screen based on the page content, freq of visit etc.
110
Szymon Kaliski @szymonkaliski.com · 25/03/2026
if you save up for 72 years these 5 extra days give you whole extra year!
000
Szymon Kaliski @szymonkaliski.com · 19/02/2026
yes! quick but simple responses just have a different feel than long and thought out ones, and I think there's design space/need for both
020
Szymon Kaliski @szymonkaliski.com · 08/02/2026
if you're on nixos: github.com/microvm-nix/...
010
Szymon Kaliski @szymonkaliski.com · 28/01/2026
I wanted to, but can't make it, I hope this will be recorded?
100
Szymon Kaliski @szymonkaliski.com · 20/01/2026
this is intriguing, curious if there's a light interface to be built around this, but what even is the right metaphor, these aren't really "filters" or "components", but there's something about the repeatability of the actions taken that feels like could be abstracted maybe..?
110
Szymon Kaliski @szymonkaliski.com · 13/01/2026
it's just the Gemini Flash model, it doesn't do anything that gemini.google.com wouldn't do, you can just paste the image there and see I guess the point is mostly the "vibe" of the interaction, not the model
020
Szymon Kaliski @szymonkaliski.com · 13/01/2026
totally! it's just an LLM after all, there's no extra magic here; if you were to draw an ASCII diagram in a chat asking about this, you'd most probably get very similar result
020
Szymon Kaliski @szymonkaliski.com · 12/01/2026
I was supposed to be a doctor you see
030
Szymon Kaliski @szymonkaliski.com · 12/01/2026
yea, sure could!
000
Szymon Kaliski @szymonkaliski.com · 12/01/2026
thank you for the explanation! I'll think about this :)
020
Szymon Kaliski @szymonkaliski.com · 12/01/2026
after the next model response, the previous live examples are frozen in time, only the latest one remains interactive (as you can draw on top of them, so if you could continue to interact, the drawing/state would drift and be confusing) -- this is more of a design choice
110
Szymon Kaliski @szymonkaliski.com · 12/01/2026
🙃 dynamicland.org/2024/FAQ/#Wh...
120
Szymon Kaliski @szymonkaliski.com · 12/01/2026
no, this generates js canvas code I like your use-case though! can you give me some examples of the kinds of diagrams you're thinking about?
150
Szymon Kaliski @szymonkaliski.com · 12/01/2026
overall the tradeoff of "quality" vs "speed" seems to make a lot of sense here, I'm just sketching something I'm interested in, and getting responses back in seconds I also really like the vibe of hand-drawn human input and "technical" model responses
0121
Szymon Kaliski @szymonkaliski.com · 12/01/2026
... or doing some math visually
270
Szymon Kaliski @szymonkaliski.com · 12/01/2026
... helping with home renovation plans ...
230
Szymon Kaliski @szymonkaliski.com · 12/01/2026
... explaining electronic circuit behaviors ...
3181
Szymon Kaliski @szymonkaliski.com · 12/01/2026
prototyping co-drawing with Gemini Flash 3 at Google in these demos "thinking" is disabled, which makes the model return tokens very quickly (all videos are realtime), and I find these rapid responses pretty good for the use-cases I'm experimenting with, like: executing simple diagrams ...
612719
Szymon Kaliski @szymonkaliski.com · 05/01/2026
this is super exciting Adam, looking forward to following the progress!
020
Szymon Kaliski @szymonkaliski.com · 05/01/2026
✉️ Q4 2025 Newsletter — Independent Consulting, and Interfacing with LLMs szymonkaliski.com/newsletter/2... --- I'm an independent consultant as of November, currently exploring at Google Creative Lab reach out if you're interested in working together! hi@szymonkaliski.com
0163
Szymon Kaliski @szymonkaliski.com · 24/12/2025
yes! but other things that were hard are still hard (data interop, "substrate" where all my apps/dynamic docs live, history tracking not for nerds, "blast radius" of making modifications and preserving old versions/data, etc!)
110
Szymon Kaliski @szymonkaliski.com · 24/12/2025
it's so weird to see this essay still pop up after seven years! @inkandswitch.com is still at it with the recent www.inkandswitch.com/malleable-so... and related work, maybe there's a more LLM-related piece brewing (dunno!) -- feels like the context both has, and hasn't, changed
120
Szymon Kaliski @szymonkaliski.com · 15/12/2025
thanks! I am familiar with that work, yes :)
220
Szymon Kaliski @szymonkaliski.com · 15/12/2025
arrows!
190
Szymon Kaliski @szymonkaliski.com · 08/12/2025
we should both see the same thing
010
Szymon Kaliski @szymonkaliski.com · 01/12/2025
is that presentation online by any chance?
120
Szymon Kaliski @szymonkaliski.com · 27/11/2025
github.com/szymonkalisk...
github.com
GitHub - szymonkaliski/name-pdfs: Auto-rename PDF files using Claude AI
Auto-rename PDF files using Claude AI. Contribute to szymonkaliski/name-pdfs development by creating an account on GitHub.
120