About this episode
News includes the public launch of Phoenix.new - Chris McCord's revolutionary AI-powered Phoenix development service with full browser IDE and remote runtime capabilities, Ecto v3.13 release featuring the new transact/1 function and built-in JSON support, Nx v0.10 with improved documentation and NumPy comparisons, Phoenix 1.8 getting official security documentation covering OWASP Top 10 vulnerabilities, Zach Daniel's new "evals" package for testing AI language model performance, and ElixirConf US speaker announcements with keynotes from José Valim and Chris McCord. Saša Juri? shares his comprehensive thoughts on Elixir project organization and structure, Sentry's Elixir SDK v11.x adding OpenTelemetry-based tracing support, and more! Then we dive deep with Chris McCord himself for an exclusive interview about his newly launched phoenix.new service, exploring how AI-powered code generation is bringing Phoenix applications to people from outside the community. We dig into the technology behind the remote runtime and what it means for the future of rapid prototyping in Elixir.
Show Notes online - http://podcast.thinkingelixir.com/259
Elixir Community News
https://www.honeybadger.io/ – Honeybadger.io is sponsoring today's show! Keep your apps healthy and your customers happy with Honeybadger! It's free to get started, and setup takes less than five minutes.
https://phoenix.new/ – Chris McCord's phoenix.new project is open to the public
https://x.com/chris_mccord/status/1936068482065666083 – Phoenix.new was opened to the public - a service for building Phoenix apps with AI runtime, full browser IDE, and remote development capabilities
https://github.com/elixir-ecto/ecto – Ecto v3.13 was released with new features including transact/1, schema redaction, and built-in JSON support
https://github.com/elixir-ecto/ecto/blob/v3.13.2/CHANGELOG.md#v3132-2025-06-24 – Ecto v3.13 changelog with detailed list of new features and improvements
https://github.com/elixir-nx/nx – Nx v0.10 was released with documentation improvements and floating-point precision enhancements