Peng Cheng, asking λP2-λC @tribbloid Twitter profile | Pikagi

Pikagi

Peng Cheng, asking λP2-λC

@tribbloid

266

Followers

276

Following

124

Media

1,510

Statuses

doubling down on Scala ♨️ and Apache Spark 🌟 for ML ∇, provers ∀ & swarms 🐝. maintainer of shapesafe, spookystuff and splain plugin

Toronto, Ontario

https://t.co/DIytP6zQrk

Joined July 2014

Don't wanna be here? Send us removal request.

Pinned Tweet

@tribbloid

Peng Cheng, asking λP2-λC

1 year

Doubling my bounty for the grand challenge: prove/refute conjecture CoC vs system-λP2 Can simplify higher-kind impl for all kinds of compilers ( #scala #rust #haskell ) and pave the way for #LLMs to learn to prove theorems after latching on them

Tweet card media

In intuitionistic type theory, can any proof written in CoC be rewritten in system λP2? Or, does...

(This question is under a permanent bounty of 1000 points, once proven/refuted, it will be retrospectively set up and awarded) (Possible duplicate: https://math.stackexchange.com/questions/4232108/...

stackoverflow.com

1

0

8

Last Seen Profiles

@Nikezy_

@doremifaso64

@JulioMacielDF

@cuckolding62764

@_Moon_Moon

@shdo07003012

@_joshuzi

@ImDaJasperBaby

@Khan_Specs

@djlegacy

@stw46

@Jukki_BatRat

@EaglesHour

@iler_ences

@stwmaniax

@bokeplokalmalam

@sempakemak05

@Servier

@Lalabhaipatel

@amckrn

@GooglePlayBiz

@sbviolin

@dxb_rf

@doremifaso64

@abflipse

@cristocurrensi

@Racckitss

@thomas_drrer

@fdsAfonso

@mindyshak

@yelmjouie

@humbertnavasg

@yupsunoo

@to_t308

@9T4EIL

@iniufill

@tribbloid

Peng Cheng, asking λP2-λC

1 month

@timClicks why do you think NVidia's advantage is MatMul hardware? it was always versatile compiler pipelines

3

2

177

@tribbloid

Peng Cheng, asking λP2-λC

5 months

you know it is CODE RED when Microsoft Edge keep involuntarily changing its default search engine from @perplexity_ai back to Bing🤨🤨

Tweet media one

3

2

27

@tribbloid

Peng Cheng, asking λP2-λC

1 month

@timClicks yeah but market valuation and control of supply chain correlates weakly. hypothetically, Blizzard built world editor and built WarCraft 3 on top of it that sells. But if the market turn to DotA, they could've retain their grip easily. (they eventually didn't, sad story)

3

0

26

@tribbloid

Peng Cheng, asking λP2-λC

2 years

@MandaloreDeFi @elonmusk must be in a yoga position

0

0

17

@tribbloid

Peng Cheng, asking λP2-λC

2 years

One committer at scalacenter and myself were quite surprised yesterday to learn that most #Scala developers still favors #IntelliJ 's unreliable Scala plugin, instead of their state-of-the-art metals language server. How did this happen?

27

0

17

@tribbloid

Peng Cheng, asking λP2-λC

1 month

@timClicks what about ray tracing? or physical simulation in Isaac?

1

1

17

@tribbloid

Peng Cheng, asking λP2-λC

3 months

just finished the seminar hosted by @aniervs et al. @ @CohereForAI 🤯all I can say is that differentiable programming is dead, long live functional & axiomatic programming Will gradually post my other impressions here🏗️🧵

@CohereForAI

Cohere For AI

3 months

Our community-led ML Theory Group is looking forward to hosting @PetarV_93 , Research Scientist at @GoogleDeepMind , next week on Thursday, April 25th for a presentation on "Categorical Deep Learning. An Algebraic Theory of Architectures." Learn more:

Tweet media one

3

12

54

0

2

17

@tribbloid

Peng Cheng, asking λP2-λC

9 months

@alxcharlesdukes @JoeyMannarinoUS "roman phalanx"? un-nerdy man, un-nerdy

2

0

15

@tribbloid

Peng Cheng, asking λP2-λC

2 years

@Grady_Booch "Nothing, unless it might be - unless it is permitted to ask, nay, to name a single strand of your hair, which surpasses the gold of the earth as the stars surpass the gems of the mine. I do not ask for such a gift"

0

0

14

@tribbloid

Peng Cheng, asking λP2-λC

1 year

@johnjnay looks promising ... how long before Copilot is allowed to interact with local compilers?

0

0

12

@tribbloid

Peng Cheng, asking λP2-λC

2 years

@aschmelyun @joewardpr @calebporzio OMG people are just immature: McDonalds #140 Lol, nice idea with the printer #141 Cool printer #142

1

0

10

@tribbloid

Peng Cheng, asking λP2-λC

5 months

@VictorTaelin there is another possibility: people used to delay it to the last stage because the sunk cost of formalising the wrong theory is too high. if it is low enough (with machine assisting), perhaps people will adapt to quickly experimenting on wrong axioms?

1

0

11

@tribbloid

Peng Cheng, asking λP2-λC

6 months

@miniapeur Jensen's inequality SGD on the other hand ...

0

0

10

@tribbloid

Peng Cheng, asking λP2-λC

1 year

All bugfixes in the latest #splain plugin has been backported to #scala 2.13.11 compiler, revealing every detail of its implicit deduction algorithm. Enabled by `-VImplicits` compiler argument @SethTisue have I told you I'll go very far for you :)

Tweet media one

1

2

10

@tribbloid

Peng Cheng, asking λP2-λC

4 months

@AISafetyMemes that's not how things work, you mostly likely need a paradigm shift to sidestep a plateau

Tweet media one

1

0

9

@tribbloid

Peng Cheng, asking λP2-λC

5 months

@Krever01 don't think so, it was only "dreaded and despised" by some random monad bro real functional programmers love it and hail it as one of their finest examples

0

0

9

@tribbloid

Peng Cheng, asking λP2-λC

7 months

@mathisstillfun a lot of people here already knew the answer by using rotation symmetry in their head, and why is it in English?

1

0

7

@tribbloid

Peng Cheng, asking λP2-λC

2 years

Announcing #shapesafe 0.1.4, a state-of-the-art static prover/type-checker for array programming in #Scala This will be the last patch for 0.1.x - the next release will focus on Scala 3 upgrade and prover efficiency. 0/1

GitHub - tribbloid/shapesafe: SHAPE/S∀F∃: static prover/type-checker for N-D array programming in...

SHAPE/S∀F∃: static prover/type-checker for N-D array programming in Scala, a use case of intuitionistic type theory - tribbloid/shapesafe

2

3

8

@tribbloid

Peng Cheng, asking λP2-λC

4 months

@AISafetyMemes hold your horse, 13%~something still has a long way to go

3

0

6

@tribbloid

Peng Cheng, asking λP2-λC

1 year

Evolution from System F ( #Haskell , #OCaml ) to DOT logic ( #Scala ) in 2 diagrams left - Hasse | right - linked hierarchy every extension attempt takes sweat, grit and over 1k lines of machine-assisted soundness proof

Tweet media one

Tweet media two

2

2

6

@tribbloid

Peng Cheng, asking λP2-λC

9 months

@yi_ding looks like uploading docs, not RAG exactly

1

0

7

@tribbloid

Peng Cheng, asking λP2-λC

5 months

@jimkxa C++ is also a swamp, but have we got away?

1

0

7

@tribbloid

Peng Cheng, asking λP2-λC

1 year

@mmbronstein oh what do you think caused the volcano eruption?

1

0

6

@tribbloid

Peng Cheng, asking λP2-λC

6 months

@mmbronstein those and "overfitting"

0

0

6

@tribbloid

Peng Cheng, asking λP2-λC

2 months

Who is using #Scala 3 for Apache Spark pipelines? Is it stable?

1

0

6

@tribbloid

Peng Cheng, asking λP2-λC

7 months

i'd like to introduce the latest OCaml feature: depened ttyes! ()

Tweet media one

2

0

5

@tribbloid

Peng Cheng, asking λP2-λC

1 year

Functional programmers ( @austinvhuang @tscholak et al.) Time to claim your slice of 🥧

Tweet media one

2

0

5

@tribbloid

Peng Cheng, asking λP2-λC

3 months

@channingwalton just want to remind you that catgrad () is also in direct-style

Tweet card media

GitHub - statusfailed/catgrad: a categorical deep learning compiler

a categorical deep learning compiler. Contribute to statusfailed/catgrad development by creating an account on GitHub.

1

0

5

@tribbloid

Peng Cheng, asking λP2-λC

5 months

Fun fact: this important article is co-authored by 2 Databricks executives and 1 former director of ScalaCentre at EPFL yet I feel that their opinions are often shadowed in the steering of #scala compiler design, what may have caused this?

@matei_zaharia

Matei Zaharia

5 months

Interesting trend in AI: the best results are increasingly obtained by compound systems, not monolithic models. AlphaCode, ChatGPT+, Gemini are examples. In this post, we discuss why this is and emerging research on designing & optimizing such systems.

30

263

1K

1

0

4

@tribbloid

Peng Cheng, asking λP2-λC

6 months

@elonmusk @MarkovMagnifico done right, a compiler should figure out program and proof from types and nothing else and the program can be much longer, do you still prefer to write it?

1

0

5

@tribbloid

Peng Cheng, asking λP2-λC

3 months

@bronzeagepapi that's an illusion of choice: physics is built on symmetry which is a monad of (I forgot the rest)

1

0

5

@tribbloid

Peng Cheng, asking λP2-λC

4 months

@fbrasisil you'll love it. IMHO it works best in CodeGen & divide-&-conquer with tools at this moment Scala community simplify don't have enough tools designed for machines, hope it will change soon

1

0

5

@tribbloid

Peng Cheng, asking λP2-λC

2 years

#shapesafe v0.1.1 for #scala is out for open beta Yet another static type-checker for N-D array programming, but armed with a mini theorem prover. It could be the first of its kind to verify einsum & Kronecker product. If #dex or #hasktorch is too heavyweight for you ...

Tweet media one

4

0

5

@tribbloid

Peng Cheng, asking λP2-λC

4 months

@fbrasisil sorry I missed your talk, will watch ASAP the Scala community is many things, certainly can't be represented by 1 sample. Fortunately, most of them can admire an unique simplification for some solid use cases

1

0

5

@tribbloid

Peng Cheng, asking λP2-λC

3 months

most #scala programmers will tell you that it means "scalable language" but do you know it's also a Latin word for "ladder"? namely, the ladder to delve into dependent type theory

Tweet media one

0

0

5

@tribbloid

Peng Cheng, asking λP2-λC

5 months

@fbrasisil voted keep: features that depends on v3 staging (direct style mostly) can become their own submodule with no v2 counterparts: many v2 users found them controversial meanwhile, Apache #Spark & CHISEL are dominant in their play field, are you ready to leave them behind?

2

0

4

@tribbloid

Peng Cheng, asking λP2-λC

5 months

@swlkr starting with augmenting cargo seems legit at least, donating instead of shovelling PR displays humility

0

0

4

@tribbloid

Peng Cheng, asking λP2-λC

4 months

@moyix NGL that's a great idea, you should submit it:

Tweet card media

Scala Contributors

for Scala contributions, language evolution discussions, standard library evolution discussions, and so on

contributors.scala-lang.org

1

0

3

@tribbloid

Peng Cheng, asking λP2-λC

5 months

@luke_metro GC is also memory safe

0

0

4

@tribbloid

Peng Cheng, asking λP2-λC

29 days

lol prompt engineers are automating themselves into extinction

@thomasahle

Thomas Ahle

29 days

Exciting new paper: TextGrad: Automatic “Differentiation” via Text! This is a DSPy-like framework for optimizing prompts in a composite LLM system. However, there is one major difference! In DSPy the idea is (basically): - Forward pass: Each Module (LLM call) generates picks a

Tweet media one

Tweet media two

Tweet media three

16

63

305

1

1

5

@tribbloid

Peng Cheng, asking λP2-λC

2 years

@Dirque_L German could use camel case

1

0

4

@tribbloid

Peng Cheng, asking λP2-λC

10 months

Latest splain plugin for #scala now actively sidesteps `invalid X <:< X` errors.

Tweet media one

0

0

4

@tribbloid

Peng Cheng, asking λP2-λC

9 months

@SokobanHero @batwood011 - diamond transitive dependencies (most package managers had this problem) - low-level C++ binaries that cannot be rewritten or linked (this is exclusive to python) - and finally package manager has GIL (!)

0

0

4

@tribbloid

Peng Cheng, asking λP2-λC

5 months

@Krever01 we've been there

Tweet card media

Implementing Higher-Kinded Types in Dotty

Implementing Higher-Kinded Types in Dotty - Download as a PDF or view online for free

www.slideshare.net

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

1 year

@DimitrisPapail it's not external if it's on the same machine or process

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

8 months

@SchmidhuberAI ... and in WH40K everybody loves necron

0

0

1

@tribbloid

Peng Cheng, asking λP2-λC

28 days

@KamerynJW oh you spent a few minutes on man page? prove the necessary condition of 2 branches being able to be merged without conflict

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

9 months

compiler designers: what stopped you from emitting errors like this? #LeanLang

Tweet card media

Paperproof: October 2023 version

www.youtube.com

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

3 months

@0xcharlie but is the one sneaking in being sued? no? then they will keep coming, and they are not necessarily human

0

0

2

@tribbloid

Peng Cheng, asking λP2-λC

8 months

me using GitHub copilot for theorem proving🤦🤦

@docmilanfar

Peyman Milanfar

8 months

Perfect

Tweet media one

7

108

849

1

0

3

@tribbloid

Peng Cheng, asking λP2-λC

7 months

@hmemcpy @lukasz_bialy speaking as a splain maintainer: most features :) but not all (e.g. showing diverging implicits & type reduction etc.) Just tests your example on my build and it looks fairly reasonable.

Tweet media one

1

0

3

@tribbloid

Peng Cheng, asking λP2-λC

9 months

@pmddomingos Bengio & students are very active

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

1 month

@UaithnesParadox @VictorTaelin in this context it should mean "operator fusion", namely aggregating subroutines with similar computation and time consumption to use 1 instruction on SIMD FPGA can fuse natively and has been exploited in

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

7 months

@tunguz i've heard in herbal medicine that it boosts male performance, haven't tested tho

1

0

3

@tribbloid

Peng Cheng, asking λP2-λC

9 months

@ekzhang1 I took his Udacity course once, many students remember him as practical, unassuming and playful but this is still gold-tier meme material😅(+KittyHawk)

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

2 years

@_JamesWard I'm sorry but the idea of mono-repo for indie teams are just unrealistic, so Gradle will still suck until Java fully embraces the principal of reproducible package (like NixOS) It is mostly Java's flaw tho

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

23 days

@VictorTaelin which safe rule? jacrev is "Jacobian accumulation in reverse order" in differentiable programs: reducing an IN of Jacobian MatMuls from its output (usually just a vector or small batch of vectors), it's very wasteful on autoencoder, and it's the lamest thing ever

2

0

3

@tribbloid

Peng Cheng, asking λP2-λC

9 months

@karpathy training on inference architecture? NVidia: :-]

0

0

0

@tribbloid

Peng Cheng, asking λP2-λC

9 months

@ChShersh indeed, but what you see is the outcome of a successful, machine-assisted campaign, which relies very little on community Devs only want to build things with least drama, not brotherhood or privileges

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

6 months

@paul_snively I often wonder how many billion-dollar mistakes are still in mainstream compilers. Was multiple inheritance deliberately designed to be non-commutative from the beginning?

2

0

3

@tribbloid

Peng Cheng, asking λP2-λC

3 months

@hmemcpy maybe in terms of correctness, but probably matters more in runtime optimisation (particularly for non-von-Neumann architecture) I prefer Odersky's summary: it is for steps in space, while imperative ones are for steps in time

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

4 months

// the implementation is left to the machine god as an exercise

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

7 months

@whatilearnedonx that was according to Gregory Bar Hebraeus, 600 years later

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

1 month

@lukasz_bialy @guizmaii "attach generated code as comment" could be a scalafix plugin feature

1

1

3

@tribbloid

Peng Cheng, asking λP2-λC

8 months

honoured drone engineers: what is preventing you from compiling ArduPilotMega on RISC-V?

@dsp8bit

dsp8bit

8 months

RISC-V + FPGA with the beaglebone ecosystem.

0

2

19

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

11 months

@PalmerLuckey lol you can always add more weight to make it 0

0

0

1

@tribbloid

Peng Cheng, asking λP2-λC

1 year

@vboykis wanna try Scala?

GitHub - tribbloid/shapesafe: SHAPE/S∀F∃: static prover/type-checker for N-D array programming in...

SHAPE/S∀F∃: static prover/type-checker for N-D array programming in Scala, a use case of intuitionistic type theory - tribbloid/shapesafe

1

0

3

@tribbloid

Peng Cheng, asking λP2-λC

11 months

@chr1sa problem: "blocked most commercial use" > 250g solution: don't go that 'eavy

0

0

0

@tribbloid

Peng Cheng, asking λP2-λC

1 year

@DrJimFan THE best? what percentage of theorems is it able to express? At this moment, the record holder should be Coq or LEAN

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

4 months

unpopular opinion: how about generating an entire #Scala collection library WITHOUT human? (model is claude3 opus)

Tweet media one

1

0

2

@tribbloid

Peng Cheng, asking λP2-λC

9 months

@matei_zaharia congrats! long overdue

0

0

0

@tribbloid

Peng Cheng, asking λP2-λC

3 months

@tangled_zans all but AI industry :D

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

1 month

@VictorTaelin meanwhile, is IC strongly normalising, even after combining with other optimisation axioms? (doesn't affect usability much, nothing EqSat can't handle)

1

0

2

@tribbloid

Peng Cheng, asking λP2-λC

2 years

@OriolVinyalsML @GaryMarcus thoughts?

1

0

3

@tribbloid

Peng Cheng, asking λP2-λC

1 year

@locallycompact Someone I can't recall summarised it very well: > Remember computers are built with bare metal, not category theory As a result, it is sullied by compromises (e.g. unlawful functors)

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

1 year

@dythui @breandan Not ignoring, approximating :D And progress is relative, most SOTA LLMs are still bad at writing strongly-typed code, the stronger the worse

@alexgraveley

Alex Graveley

1 year

As best anyone can tell, advanced reasoning in LLMs comes from learning code. Are humans the same?

33

7

129

1

0

3

@tribbloid

Peng Cheng, asking λP2-λC

1 month

@fbrasisil @alexelcu @propensive maybe it is also about existing compiler optimisations and data compression I vaguely remember that "equivalence" is defined as reconstructability / safety in substitution / no loss of info, in which case JVM equality becomes the safest "root equivalence" that lose least info

Tweet media one

0

0

1

@tribbloid

Peng Cheng, asking λP2-λC

7 months

@airkatakana apparently in Quebec "cyberpunk utopia" is a thing.

1

0

3

@tribbloid

Peng Cheng, asking λP2-λC

3 months

as a side effect: will Apache Arrow become the new MAVLink? OR, will MAVLink evolve to have shared-memory bypassing? @drlorenzmeier

@Ilir_AI

Ilir

3 months

🔥 𝗛𝗨𝗚𝗚𝗜𝗡𝗚 𝗙𝗔𝗖𝗘 🔥 - aims to replace ROS - 😱 NEW robotics framework - called Dora-rs. ⚡️A super impressive replacement of ROS (the Robot Operating System), for those who know, one of the pain-point to lower the entry barrier in robotics. 🤗Dora-rs is much much

9

38

192

1

0

3

@tribbloid

Peng Cheng, asking λP2-λC

1 year

@vzverovich Whats the type system?

0

0

3

@tribbloid

Peng Cheng, asking λP2-λC

10 months

@gdb alternatively: carelessly delegating to the machine

0

0

0

@tribbloid

Peng Cheng, asking λP2-λC

2 months

what I could never understood about average categorical programmers is the pride & feeling of adequacy in implementing easy & mundane things. While the vast scope of new capabilities often deter them: differentiable programs, composable dependency managers, hardware compilers

0

0

1

@tribbloid

Peng Cheng, asking λP2-λC

6 months

@timClicks write your own plugin in rust?

0

0

1

@tribbloid

Peng Cheng, asking λP2-λC

18 days

SIP-62 (for rewrite in #scala ) looks tempting, but do we have axiomatic rewrite rules? Can it rewrite control flow? () If such capability (e.g. folding id & untupling function at compile-time) is exclusive for a syntax sugar, something feels very wrong

Tweet card media

Pre SIP: `for` with Control Flow (an alternative to Scala Async, Scala Continuations and Scala...

Motivation Coroutines (including generators and async functions), were considered as a killer feature, but now become a usual practice, after being adopted in ECMAScript and other mainstream langua...

contributors.scala-lang.org

0

0

2

@tribbloid

Peng Cheng, asking λP2-λC

4 months

@lukasz_bialy that's a good refutation ... my hypothesis is that compilers will inevitably use globally shared resources for some accelerated empirical generation & optimisation, following examples of built tools, IDEs, AutoTVM, & Lean copilot. They are already leaking tons of information.

0

0

0

@tribbloid

Peng Cheng, asking λP2-λC

8 months

Dr Calvin Moes, lead test pilot of Toronto Human Powered Vehicle Design Team and winner of the 2012 Igor I. Sikorsky prize, evaluating a lightweight #HMD prototype with see-through capability running on a pair of #XREAL AR glasses The system weighs under 450g and is designed for

1

0

2

@tribbloid

Peng Cheng, asking λP2-λC

3 months

human should focus on the 1%

@PR0GRAMMERHUM0R

Programmer Humor

@PR0GRAMMERHUM0R

3 months

youKnowBesidesAllTheOtherStuff

Tweet media one

12

30

479

0

0

2

@tribbloid

Peng Cheng, asking λP2-λC

2 months

@Robotbeat @lightleviathan1 @yacineMTB Ondsel has a language-based intermediate representation? Since when?

2

0

2

@tribbloid

Peng Cheng, asking λP2-λC

4 months

@harris_edouard haven't read the full text but appreciate the interviews, would totally like to hear your stories in Toronto! so far, I have never known an ML engineer/scientist who is coerced/intimidated to stay in the field or be silent, I would be happy to know exceptions

0

0

1

@tribbloid

Peng Cheng, asking λP2-λC

3 months

@ezyang I doubt effect system research can be active on python/jax, most laws rely on a dependent type system which doesn't even exist there if you'd like to go beyond, scala community has @djspiewak @rssh1 @fbrasisil , and obviously, @GuannanWei and his prof @tiarkrompf

0

0

2

@tribbloid

Peng Cheng, asking λP2-λC

1 year

@tangled_zans by extension, writing this wall must be his code uglifier's job :)

0

0

2

@tribbloid

Peng Cheng, asking λP2-λC

5 months

∀ programmers who refuse to use code generating AIs: not only are you missing out, you are also ignoring the biggest advantage of type inference, streaming parsers & integrated auto-completion in hindsight, building them for human was a mistake

0

0

2

@tribbloid

Peng Cheng, asking λP2-λC

4 months

@fbrasisil @iravid_ sorry to hear that, just want to say that if you let'em staying between you and your audience, they won next time should run them through like a door

1

0

2

@tribbloid

Peng Cheng, asking λP2-λC

5 months

@fbrasisil wow, I'm clearly on the losing side :)

0

0

2

@tribbloid

Peng Cheng, asking λP2-λC

2 years

@soebrunk @PyTorch @scala_lang Nice work! Would be glad to try for industrial adoption Can't find the project, could you post a link?

1

0

2

@tribbloid

Peng Cheng, asking λP2-λC

2 months

@fbrasisil @algoraio should be done long time ago, would be interested to contribute to any neural network / optimisation usage!

1

0

2

@tribbloid

Peng Cheng, asking λP2-λC

1 year

Source:

GitHub - TiarkRompf/minidot: Dependent Object Types (DOT), bottom up

Dependent Object Types (DOT), bottom up. Contribute to TiarkRompf/minidot development by creating an account on GitHub.

0

1

2

@tribbloid

Peng Cheng, asking λP2-λC

6 months

@the_engi_nerd @luke_metro getting 25% of the accidents blamed on you?

0

0

1

@tribbloid

Peng Cheng, asking λP2-λC

3 months

@hmemcpy anti-hallucination system :D

0

0

2