Dipesh
LightNews LightNews
dipeshhh.bsky.social
Dipesh
@dipeshhh.bsky.social
74 followers 310 following 20 posts
Interested in Programming Languages and Systems.
Posts Replies Media Videos
Reposted by Dipesh
ncrenic.bsky.social
Methodology
November 18, 2025 at 9:47 AM
Reposted by Dipesh
whitequark.org
please enjoy: my Wasm-hosted, Wasm-targeting build of Clang/Clang++/LLD: a self-contained, 25 MiB (gzipped) pure function
www.npmjs.com/package/@yow...
October 26, 2025 at 9:49 AM
Reposted by Dipesh
lawrpaulson.bsky.social
New on my blog: Everything you know is wrong
lawrencecpaulson.github.io/2025/09/20/W...
Everything you know is wrong
lawrencecpaulson.github.io
September 20, 2025 at 10:43 AM
Reposted by Dipesh
kirancodes.me
New blog post~ Why Lean 4 replaced OCaml as my Primary Language

kirancodes.me/posts/log-oc...
Screenshot of the title page for the linked blog post titled "Why Lean 4 replaced OCaml as my Primary Language". The rest of the screenshot shows the first few paragraphs of the article.
August 14, 2025 at 8:59 PM
Reposted by Dipesh
priver.dev
just published "about ai"

priver.dev/blog/ai/abou...
About AI
My thoughts about AI and software engineering
priver.dev
August 5, 2025 at 12:22 PM
Reposted by Dipesh
danabra.mov
another nice lllm learning technique: take something i struggle with, ask it to create a minimal example demonstrating the same issue for me to practice, then after i figure it out, ask it to make a slightly more complex one etc. still requires a lot of direction from my side but genuinely helpful
July 18, 2025 at 8:05 PM
Reposted by Dipesh
danabra.mov
i really love lean and this may have been the most frustrating part so far. there are some great resources but they feel a bit too scattered, and the official ones are sometimes lacking. or you just can't find stuff. or it's actually for lean 3.

on the other hand the zulip community is amazing
July 20, 2025 at 1:07 AM
Reposted by Dipesh
shreyascodes.tech
so where would one start if they have a simple language ast and wants to do type checking and inference?
July 16, 2025 at 4:16 PM
Reposted by Dipesh
jepsen.mastodon.jepsen.io.ap.brid.gy
Antithesis and Jepsen are releasing a glossary of terms useful in distributed systems testing: https://antithesis.com/resources/reliability_glossary/
A distributed systems reliability glossary
A list of key concepts for building and testing reliable distributed systems, with basic definitions and deep references.
antithesis.com
July 15, 2025 at 2:45 PM
Reposted by Dipesh
danabra.mov
i love that mathematics lets you create values out of thin air. like pulling a rabbit out of the hat
/-- **The axiom of choice**. `Nonempty α` is a proof that `α` has an element, but the element itself is erased. The axiom `choice` supplies a particular element of `α` given only this proof. The textbook axiom of choice normally makes a family of choices all at once, but that is implied from this formulation, because if `α : ι → Type` is a family of types and `h : ∀ i, Nonempty (α i)` is a proof that they are all nonempty, then `fun i => Classical.choice (h i) : ∀ i, α i` is a family of chosen elements. This is actually a bit stronger than the ZFC choice axiom; this is sometimes called "[global choice](https://en.wikipedia.org/wiki/Axiom_of_global_choice)". In Lean, we use the axiom of choice to derive the law of excluded middle (see `Classical.em`), so it will often show up in axiom listings where you may not expect. You can use `#print axioms my_thm` to find out if a given theorem depends on this or other axioms. This axiom can be used to construct "data", but obviously there is no algorithm to compute it, so Lean will require you to mark any definition that would involve executing `Classical.choice` or other axioms as `noncomputable`, and will not produce any executable code for such definitions. -/ axiom Classical.choice {α : Sort u} : Nonempty α → α
July 10, 2025 at 11:39 PM
Reposted by Dipesh
lean-lang.org
In the continuing saga of surprising things users do: This post over on LinkedIn features a custom made #LeanLang environment intended to replace a Jupyter Notebook, because, you know - why not? 👀

www.linkedin.com/posts/philip...

(With apologies to @jupyter.org ❤️)
[New Blog Post] Doing Lean Dirty: Lean as a Jupyter Notebook Replacement https://lnkd.in/ejy9hE5z #lean4 | Philip Zucker
[New Blog Post] Doing Lean Dirty: Lean as a Jupyter Notebook Replacement https://lnkd.in/ejy9hE5z #lean4
www.linkedin.com
July 8, 2025 at 9:43 PM
Reposted by Dipesh
lean-lang.org
This is an... unexpected use of the #LeanLang InfoView: unnamed.website/posts/bad-ap...

But we love the creativity!

#LeanProver #DevTools #Metaprogramming
“Bad Apple!!” But It’s 3288 Lean Tactics Spamming VSCode
Writing the most useless Lean tactic ever
unnamed.website
July 8, 2025 at 8:47 PM
Reposted by Dipesh
kirancodes.me
PSA! Please share around! Due to a limited number of submissions, we're extending the OCaml Workshop deadline by a week to July 10th AoE!

Functional programmers! Heed my call! We need your submissions!!
kirancodes.me Kiran (She/Her) @kirancodes.me · Mar 25
Calling for Presentations!!!

The OCaml Workshop 2025 welcomes presentations on any topic related to OCaml (such as multicore, algebraic effects, testing, ppxs, etc.) Have fun!!

Submissions due: July 3rd AoE
Workshop: Oct 17th

Here's the link to website:
conf.researchr.org/home/icfp-sp...

#OCaml
OCaml 2025 - OCaml Users and Developers Workshop 2025 - ICFP/SPLASH 2025
The OCaml Users and Developers Workshop brings together industrial users of OCaml with academics and hackers who are working on extending the language, type system, and tools. Previous editions have b...
conf.researchr.org
July 6, 2025 at 7:59 AM
Reposted by Dipesh
welltypedwit.ch
yeah, it's called C
June 29, 2025 at 6:15 PM
Reposted by Dipesh
eikopf.com
has someone made a programming language where all programs are UB? that feels like low-hanging esolang fruit
June 29, 2025 at 6:14 PM
Reposted by Dipesh
sigplan-pldi.bsky.social
Another award that is announced at PLDI is the SIGPLAN Programming Languages Software Award. This year, the award was received by the LEAN theorem prover, which has had and continues to have broad impact on mathematics, hardware and software verification, and AI!
June 28, 2025 at 4:04 AM
Reposted by Dipesh
kirancodes.me
If anyone has any slow or brittle Dafny/Boogie/Viper proofs, consider hiring me *hint* *hint* *nudge* *nudge* *wink* *wink*

(I would normally be asking in person while presenting this work at CAV but again... I can't leave the country right now.)
kirancodes.me Kiran (She/Her) @kirancodes.me · Jun 8
New blog post! The Looming Problem of Slow & Brittle Proofs in SMT Verification (and a Step Towards Solving It)

kirancodes.me/posts/log-pr...

#ProgrammingLanguages #Dafny #SMT #Z3 #Verification
Screenshot of the landing page of my website on the new blog post; The title reads "The Looming Problem of Slow & Brittle Proofs in SMT Verification (and a Step Towards Solving It)". The contents of the post read "All is not well in the land of automated formal verification." And then goes into the body of the article. There is a snippet of dafny code present on screen.
June 17, 2025 at 4:18 PM
Reposted by Dipesh
danabra.mov
yea honestly this video is fantastic, very inspiring in terms of how to teach things. it’s much more intuitive than most introductory neural network materials
olafinho.xyz olafinho.xyz @olafinho.xyz · Jun 15
Yeah, for soft launch the Karpathy series with no priors, I'd start here www.youtube.com/watch?v=VMj-...
also I'd pair it with maybe some written material focusing more on giving 'higher level' picture of things
bsky.app/profile/olaf...
June 15, 2025 at 6:51 PM
Reposted by Dipesh
danabra.mov
has anyone learned AI fundamentals recently? i don’t mean like “using” AI but more like, actual ML fundamentals, Transformers, whatever the underlying mechanism up to modern LLMs. including math and basic insights. i’m wondering if it’s Fun To Learn and whether it’s Useful. and fav resources if any
June 11, 2025 at 1:10 PM
Reposted by Dipesh
hillelwayne.com
n² likes = n stupid Python tricks
April 2, 2025 at 2:30 PM
Reposted by Dipesh
danabra.mov
what’s your favorite unhinged programming language feature? something that seemed ok to whoever designed it but is truly the stuff of nightmares
June 14, 2025 at 9:50 PM
Reposted by Dipesh
thenumb.at
We've been working on this for years 📈
yminsky.bsky.social Yaron Minsky @yminsky.bsky.social · Jun 13
I'm pleased to announce OxCaml!

OxCaml is Jane Street's branch of OCaml. We've given it a new name and a snazzy logo, and done a bunch of work to make it easy for people to try.
June 13, 2025 at 2:15 PM
Reposted by Dipesh
lesbianmonad.com
yes you can have the (current state of the) code
some parser combinator draft
some parser combinator draft. GitHub Gist: instantly share code, notes, and snippets.
gist.github.com
June 13, 2025 at 2:39 PM
Reposted by Dipesh
yminsky.bsky.social
And here's a blog post, announcing the release!

blog.janestreet.com/introducing-...
Introducing OxCaml
At Jane Street, we’ve been actively making improvements to OCaml for a long time. Over thelast few years, we’ve started to build some fairly ambitious extens...
blog.janestreet.com
June 13, 2025 at 2:22 PM
Reposted by Dipesh
frankmcsherry.bsky.social
A new blog post on an interactive Datalog in Rust! Early days, but 1. nice and interactive, and 2. not less performant than my prior attempt (datafrog) when you use strings as the literal type (vs u32).

More to come, as it's partially meant as a playground for me.

github.com/frankmcsherr...
github.com
June 13, 2025 at 10:11 PM