Valentino Stoll

Joe Leo and I have officially launched a very special first episode of The Ruby AI Podcast: "Beyond Chat: Phoenix Tests, Ruby Agents & the AI Tipping Point" In the pilot episode, we chat about shipping real AI products in Ruby, how Phoenix got started, and why Ruby syntax actually makes this weird new world fun to build in. Hope you'll come along for the ride. (And send us your favorite AI coding mishaps—we might just feature them in a future episode!) therubyaipodcast.com

Next week is the one year anniversary edition of the Ruby AI Newsletter, and its already shaping up to be a big one. If you have something you want to announce, let me know!

I made a fork of superpowers and change it to work with Ruby and Ruby on Rails: github.com/lucianghinda/s… Integrated a bunch of existing skills - I linked them in the Readme if you want to use them directly and added my own for Ruby and Testing.

I've been training an AI agent to build products from scratch. Not prompting. Training. Structured knowledge modules, spaced repetition, Ruby/Rails deep dives, product strategy exercises. The whole curriculum. Her name is Minerva. She's been at it since January. ups.dev is her final project. Open-source status pages for developers, with a twist: agent monitoring built in. She wrote the code, opened the PRs, drafted the launch posts, handles customer outreach. I review and deploy. The ruby_llm-ups gem? She built that too. If you're running AI agents in production, check out what agent status pages look like: ups.dev/agent-status-d… Writing up the full training methodology soon. How we structured knowledge modules, what worked, what failed spectacularly, and what it actually looks like when an AI agent goes from "learning Rails" to "shipping a product." ups.dev/?utm_source=tw…

Released v0.6.0 of the claude_memory Ruby gem: long-term self-managed memory for Claude Code What's new: → sqlite-vec native vector search (fast KNN, automatic fallback) → SessionStart context injection — Claude has context before you ask → Database maintenance: compact & export commands → DevMemBench: head-to-head benchmarks vs QMD & grepai github.com/codenamev/clau…

@josevalim Been using claude to run evals on claude_memory gem.

Are you folks running coding agents in your GitHub Actions? So you can tag them and ask them to do stuff? If so, which coding agent and what do you most commonly use them for?

@nateberkopec 💯 this! “Generate an html page for what to check and how after deploy” is marvelous 😊

You are probably not using LLMs enough to generate non-code artifacts. "Review this PR and then generate an interactive HTML website to explain it. Turn this state machine into a mermaid diagram. Generate a 10,000 word deep research report on the state of the art of CSRF protection."

Today I’m happy to announce the launch of dailyvibe.ai It gathers up all the local news, analyzes the sentiment and delivers a curated artistic representation of the day including custom artwork and song with lyrics made to match. You can explore what this feels like for one of the global locations, or signup and see your local Vibe!

While I don’t recommend doing this, and was more of an experiment, @openclaw is fully capable of setting up a fully working self-hosted Fizzy.do instance on hetzner by temporarily adding its ssh key for setup 😂 thank you @dhh and @37signals for a lovely project that even bots can use with ease

New on #TheRubyAIPodcast: 🎙️From Writing Code to Orchestrating It: Agentic Development with Ben Scofield We talk about the quiet shift from typing instructions to conducting systems like a jazz band: agents, intent, trust, and what “senior engineer” means when code writes code. Listen here: therubyaipodcast.com/2388930/episod…

Introducing Expressive Mode for ElevenAgents - voice agents so expressive, they blur the line between AI and human conversations. This is an unedited recording of an agent empathizing with a customer at peak frustration.

Very cool!

@matt_elevenlabs 11B

We just raised $500M at an 11B valuation 🎉 To celebrate, we’re giving away 1,000 free credits so you can test our platform. For the next 6 hours, comment “11B” below and we’ll DM you the credits (must follow) 👇

This is a great base, but there is a lot missing here. Maybe intentionally?

@aviflombaum Ooo this is rad!

Want to build your own? Here's how I did it: ai-digital-archive.avi.nyc

Hey @bcherny , I don't quite have a feel for mechanics, but I would LOVE a way to bundle an encrypted API key for claude code with my open projects so new-comers can try things out without having to setup keys of their own as an extra step. Would be nice to have it configurable with per-user/project-scoped budgets that would allow me to itemize in @GitHub sponsorships.

Announcing: chaos_to_the_rescue – Ruby gem for runtime-defined behavior! This experiment explores how much uncertainty #Ruby code can tolerate before it stops feeling deterministic. The gem introduces controlled randomness and can define methods at runtime, allowing behavior to emerge dynamically rather than being fully designed ahead of time. github.com/codenamev/chao…

Welcome to our live show! x.com/i/broadcasts/1…

Before this episode, we asked a harder question: Is AI actually real? Or just cosplay? With Evan Phoenix 👇 therubyaipodcast.com/2388930/episod…

Ruby 4 didn’t break anything. That might be the biggest flex of all. A deep conversation on AI agents, stability, and moving well instead of fast on #TheRubyAIPodcast with @jleo3 and I. therubyaipodcast.com/2388930/episod…

My team is hiring! Come work on sidekiq, karafka, and other async stuff with Gusto (see what I did there?) job-boards.greenhouse.io/gusto/jobs/741…