Building infrastructure for AI agents to write code @crunchloop. Kubernetes, Linux namespaces, devcontainers, MCP. https://nitter.cf/t.co/pG0ucoHpVI
Montevideo, Uruguay
Joined April 2009
- Tweets518
- Following351
- Followers123
- Likes188
Martín Fernández retweeted
We’re hosting the Grok Bot Meet-Up in Buenos Aires 🇦🇷
Wed, Nov 4 · 6:30–10:30 PM
Working&Co, Soler 4445, CABA
Come meet the @crunchloopio team
luma.com/myheojcx
Martín Fernández retweeted
Con @crunchloopio vamos a estar hosteando la Grok Bot Meetup en Buenos Aires!
El miercoles 4/11 vamos a juntarnos para charlar sobre el producto y desde Crunchloop vamos a estar comentando como venimos aplicando las herramientas a nuestro trabajo del día a día.
Les dejo el link porque va a volar luma.com/myheojcx
Martín Fernández retweeted
We went to the ORT job fair to talk about how we work.
We ended up listening more than we talked.
That’s part of how we build a team, understand what people are looking for before telling them what we offer.
#Crunchloop #LifeAtCrunchloop #ORTUruguay #TechUruguay
Martín Fernández retweeted
Our office DJ is a Full Stack Engineer.
Fede shares the 5 radio stations that play nonstop at @crunchloopio 📻
#LifeAtCrunchloop
Martín Fernández retweeted
You don't get trust in agents by asking for it. You get it by making the reasoning visible.
The question that matters after an agent touches your code isn't "did it work?" It's "what did it do, and why?" — and if that answer lives in a log nobody reads, trust stays where it started: at zero.
Trust is built when you can see each decision, the context behind it, and the trail it left — before, during, and after the work.
That's why visibility is the product, not a feature of it. It's what we built Dap's governance layer around.
More here: crunchloop.io/platform/use-c…
Martín Fernández retweeted
Startup Day at Xplora UCEMA ✅
Ignacio and Alejandro represented @crunchloopio on September 11, talking with founders and builders about what they're working on and where software fits in.
Thanks to the Startup Day team, See you at the next one.
#StartupDay #Argentina
I don’t think most people understand how much compute usage per person is going to increase as we scale agentic workflows.
I started renting a server with 32 vCPUs, 128 GB of RAM, and NVMe storage for a development project, and my parallel workloads are already maxing it out.
Martín Fernández retweeted
Rails World 2026, done.
Two days at the booth, a lot of good conversations about engineers and agents working in one loop, and one tired team.
Thanks, Austin. More on what we learned soon.
crunchloop.io/platform/featu…
#RailsWorld #Dap #Crunchloop #RailsWorld2026
Martín Fernández retweeted
Day 1 at Rails World, done.
Dap on screen, stickers on the table, and a lot of good conversations about engineers and agents working in one loop.
crunchloop.io/platform/featu…
#RailsWorld
Martín Fernández retweeted
Something I see constantly: teams "adopt AI", but every engineer ends up running their own flavor of it.
The intent is good. The consequence isn't: DORA's 2025 research names "a clear and communicated AI stance" as one of the few things that reliably amplifies AI's benefits across a team — and their advice is blunt: "Establish and socialize a clear policy on permitted tools and usage to build developer trust."
Standardizing that across engineers is exactly what lets AI's real impact show up at the team level, instead of getting diluted across individual habits.
This is not a nice-to-have for us — it's the differentiator.
The use case: crunchloop.io/platform/use-c…
#AI #AgentOps
It’s almost time for my trip to the U.S. with the Crunchloop team! If you’ll be in Austin or Dallas and would like to chat, send me a DM.
Martín Fernández retweeted
When your platform unlocks unattended agents, your outcomes can no longer be measured in engineering hours. Curious how teams that have stepped into this territory are measuring things — especially the ones not capped by a subscription.
Agent work is about to show up as a real line item and most teams haven't prepared a budget line for it.
The instinct is to compare against a salary. The better comparison is per unit of work: token spend is a cost you can see, predict, and optimize. Teams that treat agent spend as engineering economics, measured and allocated, will make calmer decisions than teams that treat it as a surprise on an invoice.
We've been running this math on our own platform for a while. If agent work is heading for your P&L, the details are worth your attention: crunchloop.io/platform/featu… and crunchloop.io/pricing
Agent work is about to show up as a real line item and most teams haven't prepared a budget line for it.
The instinct is to compare against a salary. The better comparison is per unit of work: token spend is a cost you can see, predict, and optimize. Teams that treat agent spend as engineering economics, measured and allocated, will make calmer decisions than teams that treat it as a surprise on an invoice.
We've been running this math on our own platform for a while. If agent work is heading for your P&L, the details are worth your attention: crunchloop.io/platform/featu… and crunchloop.io/pricing
Martín Fernández retweeted
One week until #RailsWorld.
We’ll be in Austin on Sept 23–24 with a live #Dap demo at the Crunchloop booth.
Come talk about what it takes to run #AI in a real software delivery loop.
Not attending? Explore Dap: crunchloop.io/
#Rails #Crunchloop #SoftwareEngineering
Martín Fernández retweeted
El viernes pasado estuvimos presentes con @crunchloopio en el StartupDay de @xplora_ucema contando lo que venimos haciendo con @hey_im_dap
Muchas de las empresas con las que hablamos están hasta las manos desarrollando cosas y no llegando con las planificaciones que se tienen. La AI nos da el sentimiento de tener superpoderes pero si no usa bien puede generar más quilombos que beneficios.
Si alguno se siente así dejo los MD abiertos para charlar y contarles cómo lo resolvimos con DAP!
Martín Fernández retweeted
I think agents became principals. Just another developer seat that we have to account for. They scale differently, and maximizing their outcome means introducing structures or elements to the system’s architecture that weren’t that needed before (mainly because of the feedback loops). Feels weird, though
Martín Fernández retweeted
Our leverage isn't typing. It's planning and knowing the business well enough to sequence the work.
All of it idles the moment nobody's at the keyboard.
Unattended doesn't mean unsupervised. The loop keeps running; the judgment stays in it.
crunchloop.io/platform/featu…