Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Workflows distribuídos de AI com Elixir e Livebook
Search
Hugo Baraúna
September 28, 2024
Programming
0
120
Workflows distribuídos de AI com Elixir e Livebook
Hugo Baraúna
September 28, 2024
Tweet
Share
More Decks by Hugo Baraúna
See All by Hugo Baraúna
The Architecture Behind Deploying Livebook Apps with Livebook Teams
hugobarauna
0
93
Visão geral sobre Elixir @ Elixir Carajás Meetup
hugobarauna
0
19
The Architecture Behind Deploying Livebook Apps with Livebook Teams
hugobarauna
0
110
A história por trás da criação do Elixir
hugobarauna
0
36
Using Livebook to build and deploy internal tools @ ElixirConf 2024
hugobarauna
0
460
Other Decks in Programming
See All in Programming
Processing Gem ベースの、2D レトロゲームエンジンの開発
tokujiros
2
120
Swift Updates - Learn Languages 2025
koher
2
460
Flutter with Dart MCP: All You Need - 박제창 2025 I/O Extended Busan
itsmedreamwalker
0
140
HTMLの品質ってなんだっけ? “HTMLクライテリア”の設計と実践
unachang113
4
2.7k
複雑なドメインに挑む.pdf
yukisakai1225
5
1.1k
パッケージ設計の黒魔術/Kyoto.go#63
lufia
3
430
Microsoft Orleans, Daprのアクターモデルを使い効率的に開発、デプロイを行うためのSekibanの試行錯誤 / Sekiban: Exploring Efficient Development and Deployment with Microsoft Orleans and Dapr Actor Models
tomohisa
0
240
Protocol Buffersの型を超えて拡張性を得る / Beyond Protocol Buffers Types Achieving Extensibility
linyows
0
110
そのAPI、誰のため? Androidライブラリ設計における利用者目線の実践テクニック
mkeeda
2
200
基礎から学ぶ大画面対応(Learning Large-Screen Support from the Ground Up)
tomoya0x00
0
370
1から理解するWeb Push
dora1998
7
1.8k
How Android Uses Data Structures Behind The Scenes
l2hyunwoo
0
310
Featured
See All Featured
GraphQLとの向き合い方2022年版
quramy
49
14k
Why You Should Never Use an ORM
jnunemaker
PRO
59
9.5k
The MySQL Ecosystem @ GitHub 2015
samlambert
251
13k
Design and Strategy: How to Deal with People Who Don’t "Get" Design
morganepeng
131
19k
The Language of Interfaces
destraynor
161
25k
Music & Morning Musume
bryan
46
6.8k
The Myth of the Modular Monolith - Day 2 Keynote - Rails World 2024
eileencodes
26
3k
How to train your dragon (web standard)
notwaldorf
96
6.2k
I Don’t Have Time: Getting Over the Fear to Launch Your Podcast
jcasabona
33
2.4k
[RailsConf 2023 Opening Keynote] The Magic of Rails
eileencodes
30
9.6k
ピンチをチャンスに:未来をつくるプロダクトロードマップ #pmconf2020
aki_iinuma
126
53k
Rails Girls Zürich Keynote
gr2m
95
14k
Transcript
Work fl ows Distribuídos de IA com Elixir e Livebook
By Hugo Baraúna (@hugobarauna)
Functional Concurrent Distributed {
livebook-dev/livebook
Work fl ows Fine tuning de um modelo 1 Inferência
usando um modelo 2
None
“Quantas dessas 1,500,000 patentes tem relevância pra gente”
“Temos um dataset de 8,000 patentes marcadas como relevantes ou
não”
Solução Treinar um modelo classi fi cador de patentes relevantes
para o cliente
Hugo Baraúna @hugobarauna Dev advocate @ Livebook 🥑 Eng. Computação
- Poli USP 💻 Co-fundador @ Plataformatec 🖇
Live coding
Próximos passos
Install Livebook https://livebook.dev
Livebook Neural Network Smart cells https://livebook.dev
AI GPU Clusters com Livebook & Fly.io https:// fl y.io/blog/ai-gpu-clusters-from-your-laptop-livebook/
Case de AI com Elixir https://youtu.be/5FlZHkc4Mq4
Livro sobre Machine Learning em Elixir https://pragprog.com/titles/smelixir/machine-learning-in-elixir/
Código das demos https://github.com/hugobarauna/livebook-notebooks/tree/main/sorry-dave
obrigado livebook-dev/livebook livebook.dev @livebookdev @hugobarauna