José Valim on why Elixir and machine learning are a great match
A clip from "The Changelog" podcast. Full audio 👉 https://changelog.fm/538 Subscribe for more! 👇 Apple: https://changelog.fm/apple Spotify: https://changelog.fm/spotify Android: https://changelog.fm/android Overcast: https://changelog.fm/overcast Email: https://changelog.fm/email Twitter: https://twitter.com/changelog Mastodon: https://changelog.social/@changelog #software #opensource #programming #softwareengineering #coding #podcast #oss #foss #development
Sean Moriarty - The Future of Large Language Models is Elixir
Large-language models (LLMs) are leading a new wave of technological innovation. LLMs make it possible to develop and deploy increasingly complex and impressive applications. However, building LLM-powered applications also brings about a unique set of challenges and considerations. In this talk, I'll discuss the challenges of building LLM-powered applications, and why Elixir is the best choice to power the LLM-powered applications of the future. About Sean Sean is the author of the book Genetic Algorithms in Elixir: Solve Problems with Evolution. His interests include artificial intelligence, evolutionary algorithms, mathematics, and functional programming.
- cool talk describing how one can perceive LLMs as we perceive databases, like a utility rather than the main focus of your talk
- regarding non determinism in LLM outputs →
Skip the API, Ship Your Database
The best code is the code unwritten. Avoid building a complex API layer by sharing your database between services.
- small trick: use
git branch --sort=-committerdateto list your branches sorted by most recent. helpful when you’re jumping between numerous branches and forget the exact branch name
- also there’s
git checkout -to checkout to your previous branch. helpful when you want to jump to
main, take a pull, jump back to your branch and merge