Adelbert Chang

Living in the end-to-end argument @octoml , accelerating ML model deployments since 2019. 💻📚🍜🏋️‍♂️My tweets are my own.

[Adelbert Chang @adelbertchang]

🧘🏻‍♂️

[Adelbert Chang @adelbertchang]

@diptanu We used to but ended up switching to Discord.

[Adelbert Chang @adelbertchang]

@runarorama I like Callahan’s Habanero https://t.co/tJQsaeLAd6 , or you can try some of these interesting ones https://t.co/UNcybnNV7l

[LangChain @LangChain]

LangChain ❤️ @OctoML

OctoAI is world-class compute infrastructure for tuning and running models that wow your users.

We recently added integrations for LLM and embedding models on their platform!!

https://t.co/4prFBi0jkh

[Adelbert Chang @adelbertchang]

I hate ORMs

[Adelbert Chang @adelbertchang]

@lichess Qb3#, ...Rxb3 is not possible due to the pin

[Daring Fireball @daringfireball]

Bob Iger’s Letter to Disney Employees Accompanying Their Copies of ‘Make Something Wonderful’
https://t.co/lJqqPdhuKM

[Adelbert Chang @adelbertchang]

@elidourado Qb5# - Bishop can't take due to its own pin

[GothamChess @GothamChess]

Ding Liren almost didn't even play the Candidates.
Ding Liren nearly failed to qualify to play for the title.
Ding Liren trailed 3 times against Nepo in the match.

Ding Liren... is the new World Chess Champion.

[Adelbert Chang @adelbertchang]

current status

[Adelbert Chang @adelbertchang]

Kustomize makes me sad

[Adelbert Chang @adelbertchang]

I recently got a new laptop and had to setup a quick development environment for some light coding I was doing to dogfood our product. I got VSCode with remote editing + Python IntelliSense + virtual envs working in < 5 minutes. Pretty damn good onboarding UX.

[PyTorch @PyTorch]

Don't miss the live demo on @OctoML Profiler!

🗓️ Thursday, April 6 at 10 am PT
💻 Register: https://t.co/KUW9HQzhoo

[Adelbert Chang @adelbertchang]

I'm excited to share a new PyTorch tool we've been working on to help devs easily profile their PyTorch-powered applications across various HW targets! Benchmarking models and provisioning HW is annoying, but we can help - try out octoml-profile today! https://t.co/1EXY6D383f

[PyTorch @PyTorch]

@OctoML Profiler uses the latest PyTorch 2.0 compiler technology to give developers insight into ML model performance, optimize deep learning applications, and right-size hardware choices without leaving the dev environment!

Check it out 👉 https://t.co/c1hCTyOsNy

[Adelbert Chang @adelbertchang]

Independent of the philosophical arguments being made in the comments, this is still pretty cool to see.

[Adelbert Chang @adelbertchang]

Are people still using https://t.co/cHjID8EQBi? Several years back it was a more focused HN, wondering if folks still feel that holds.

[Jared Roesch @roeschinc]

Great to see some of the work our team @OctoML has done with @Microsoft in a form we can talk about it. Stay tuned for more exciting updates on how to easily access this kind of performance for your models this Spring! https://t.co/fpZ462AyhV

[Adelbert Chang @adelbertchang]

OH: You either die a Rust shop or live long enough to become a Go shop

[Adelbert Chang @adelbertchang]

Does anyone use a MacBook Air for work? How do we like it? Thinking about moving to one, given my move to Engineering Management 👀

[Adelbert Chang @adelbertchang]

A month and a half in, and I haven't looked back.

[Adelbert Chang @adelbertchang]

Realizing, especially b/c I no longer code, that I can probably replace my laptop with an iPad + Magic Keyboard.. has anyone tried this?

[Adelbert Chang @adelbertchang]

Also realized when traveling with my work laptop I can use the iPad as a second monitor with Sidecar, very nice.